Most RWE teams in pharma are fluent in US claims logic. They know Optum. They know IBM MarketScan. They know what a nationally representative claims database looks like, how to run a feasibility assessment against it, and roughly how long it takes to get from a data access agreement to first query results. They have done it before. It works.
Then they get tasked with a European launch.
And the mental model breaks — quietly, expensively, and usually at the worst possible moment in the evidence timeline.
Here’s the uncomfortable truth that most vendor decks and EMA guidance documents won’t tell you directly: there is no European equivalent of a national claims database. What exists instead is a patchwork of statutory insurance systems, hospital administrative data, national registries, EHR networks, and — increasingly — synthetic datasets that vary dramatically by country in coverage breadth, data completeness, GDPR accessibility, and analytical fit-for-purpose. Navigating this landscape as if it were a single market is one of the most common and costly mistakes in pan-European launch planning.
This post is for HEOR directors, market access leads, and RWE strategists who are planning — or already managing — evidence generation across multiple Continental European markets. It’s a practical map of what actually exists country by country, why the fragmentation happens structurally, how to design around GDPR as a real-world constraint rather than treating it as a legal veto, and how to build a sequenced evidence strategy that your national payers will actually accept.
We’ll also be honest about where the knowledge gaps are, where synthetic and aggregated data fills them, and why the companies that navigate European RWD complexity best are almost always the ones that started planning earliest.
Why European Health Data Is Structurally Different from US Claims Data
The fragmentation of European health data is not a technology problem or a regulatory failure. It is a direct consequence of how European healthcare systems are organised — and that architecture is unlikely to change materially in the short term, regardless of what the European Health Data Space (EHDS) eventually delivers.
In the US, a relatively small number of large commercial claims aggregators — built on top of employer-sponsored insurance and Medicare — have consolidated longitudinal records covering hundreds of millions of covered lives under consistent data models. The infrastructure for pan-national research exists because the insurance architecture itself is semi-centralised. A researcher can access a single database and conduct a nationally representative study with reasonable confidence that the data captures most medically relevant care events.
Europe does not have that infrastructure because European healthcare doesn’t work that way. Healthcare funding and administration is a national — and in many cases, sub-national — responsibility. Each country operates its own statutory health insurance system, hospital funding framework, and health data governance regime. These systems sit on top of GDPR as the overarching data protection framework, but GDPR is implemented differently in each member state, with national legislation adding specific restrictions — particularly for health data as a special category under Article 9. The result is that claims and administrative health data sits with different custodians in each country: statutory insurers in Germany, a centralised national insurance system in France, hospital trusts and GP networks in the UK, regional health authorities in Spain and Italy.
What this means practically for evidence generation is significant. A study that would take three months to execute in the US — from data access agreement to analysis-ready dataset — can easily take twelve to eighteen months in Continental Europe, not because the data doesn’t exist, but because accessing it requires separate national approval processes in each market, each with their own data governance requirements, ethics review pathways, and GDPR compliance documentation. Country-level data is often the only realistic option. Cross-border patient-level linkage, which would allow a single consistent analysis across multiple EU markets, is severely restricted by GDPR’s data transfer rules and by the absence of a functioning cross-border health data infrastructure — at least until EHDS secondary use provisions are operational.
Understanding this structural reality is the prerequisite for everything else in European RWE strategy. Teams that start from “how do we build one pan-European study” almost always end up with a worse evidence package than teams that start from “which country data sources are fit for purpose for which specific evidence questions, and in what sequence do we need them for our market access timeline.”
Country-by-Country: What European Claims and RWD Actually Exists
Rather than cataloguing every European data source — which would be neither practical nor stable, since the landscape evolves — what’s more useful here is a strategic tiering of markets by data richness and accessibility. This is the mental model that shapes how MEDDDICAL advises companies on evidence planning: not which vendors cover which countries, but what you should realistically expect from each market and how to sequence accordingly.
The European health insurance datasets landscape broadly stratifies into three tiers based on data maturity, accessibility, and analytical utility.
Tier 1 — Rich data, established access routes. France is, by most measures, the most powerful single-country RWD source in Continental Europe. The SNDS (Système National des Données de Santé) covers virtually the entire French population — approximately 67 million individuals — and integrates hospital discharge data, outpatient claims, prescription records, and mortality data in a longitudinal, linkable format. It is among the largest and most comprehensive administrative health databases anywhere in the world. Access requires approval from the Health Data Hub (HDH), which administers GDPR-compliant data access under French national health data legislation. For pharma evidence generation, SNDS is the reference dataset for HTA submissions to HAS (France’s national HTA authority), and payers expect to see it. The access process is procedurally rigorous but well-established — this is the market where investment in access infrastructure pays the clearest dividend.
Germany’s statutory health insurance claims — covering approximately 90% of the population through around 100 statutory health insurers — are rich in diagnostic (ICD-10 coded), prescription (ATC classified), and procedure data. However, unlike France’s centralised system, German claims data sits with individual insurers rather than a unified national database. Access routes vary: some insurers license anonymised datasets to researchers through established commercial data intermediaries; others require formal data use agreements under national health insurance data regulations. National HTA bodies evaluating new medicines have strong preferences for German-specific claims evidence, and the complexity of the German data landscape means that understanding which sources are appropriate for which evidence questions is genuinely non-trivial.
The Nordic countries — Sweden, Denmark, Finland, and Norway — collectively represent some of the most analytically powerful RWD environments in the world. Population-level registries linked across healthcare touchpoints (inpatient, outpatient, prescription, cause of death, cancer) with stable national personal identifiers enable the kind of longitudinal, multi-dimensional analysis that is simply not possible in most other European markets. The Nordic data environment is also relatively stable from a GDPR perspective, with well-established research ethics and data governance frameworks. The practical limitation is that Nordic populations are small — Sweden and Denmark each cover approximately 5 to 10 million individuals — which means statistical power for rare disease or subgroup analyses can be constrained.
Tier 2 — Accessible with navigational complexity. The Netherlands has a strong health data infrastructure through Vektis (aggregated health insurance claims covering the entire Dutch population) and Dutch Hospital Data (clinical hospital discharge records). Access is regulated and requires formal data requests, but the Dutch health data landscape is relatively transparent about what is available and how to apply. Spain presents a more complex picture: health data is managed at the regional autonomous community level, meaning that “Spanish claims data” is in practice a collection of seventeen separate regional systems with varying coverage, coding standards, and access frameworks. Catalonia has the most developed data infrastructure and is often the most accessible entry point for Spanish evidence generation, but findings from Catalonia may not extrapolate credibly to a national Spanish submission. Italy operates similarly, with regional administrative databases (in Lombardy, Veneto, Lazio) that are individually rich but structurally fragmented at the national level.
Tier 3 — Emerging infrastructure, limited current access. Central and Eastern European markets — Poland, Czech Republic, Hungary, Romania — are building health data infrastructure but are not yet at the point where systematic access for pharma evidence generation is routine. Poland has national health fund data (NFZ) that is used for health technology assessment, but external access for commercial evidence generation is more limited. These markets are often served by epidemiological data — population-level estimates derived from registry data and national statistical sources — rather than the patient-level longitudinal claims analysis possible in Tier 1 markets.
The UK deserves its own note. NHS Hospital Episode Statistics (HES) and the Clinical Practice Research Datalink (CPRD) are among the most analytically sophisticated health data environments anywhere — HES covers all NHS inpatient admissions, and CPRD provides longitudinal primary care data linked to secondary care, mortality, and cancer registry records. Post-Brexit, UK data access sits outside the GDPR framework (governed by UK GDPR and the UK Data Protection Act), and cross-EEA data transfers require specific legal mechanisms. For companies that include the UK in their European launch sequence, the data infrastructure is excellent; the governance is a separate track from Continental European access.
GDPR Is Not a Blocker — It Is a Design Constraint
Here is where most pharma RWE strategies stall — and where the most avoidable mistakes happen.
GDPR is widely understood in principle by pharma legal and compliance teams. It is poorly understood in practice by the HEOR, market access, and commercial analytics teams who are actually designing the studies and scoping the data. The gap between those two levels of understanding creates a specific failure pattern: a study is designed, a vendor is engaged, a budget is committed — and then, at the data transfer stage, a regulatory affairs or legal review flags a GDPR compliance issue that should have been designed around from the start.
The core issue for European pharma evidence generation is GDPR Article 9, which classifies health data as a special category requiring explicit legal basis for processing. This is understood. What is less consistently understood is what GDPR Articles 44–49 mean for cross-border data transfer — specifically, that transferring patient-level health data outside the EEA for processing (for example, to a US-based analytics team) requires either a formal adequacy decision, standard contractual clauses with appropriate supplementary measures, or binding corporate rules. In practice, for pharma evidence generation, this means that the analytical model designed around US-based infrastructure — where patient-level data flows freely to wherever the analytics team sits — often requires substantial architectural reconfiguration to work in a European context.
There are three practical responses to this constraint, and understanding which one applies to a given evidence question changes the study design and the data source evaluation entirely.
First, data localisation: analysis is conducted within the EEA, either through secure compute environments provided by the data custodian or through an approved research organisation that has established data residency infrastructure. This is how major national datasets like SNDS in France are typically accessed — researchers query the data through a controlled, data-residency-compliant environment rather than extracting patient-level records for external analysis.
Second, aggregated data: if patient-level linkage is not required for the evidence question, aggregated datasets — statistically representative but stripped of individual-level records — are GDPR-compliant without requiring the same legal infrastructure for transfer. Many European disease burden analyses, epidemiological estimates, and population-level treatment pattern analyses can be executed on aggregated data with appropriate methodology. The limitation is that aggregated data cannot support certain endpoint analyses — patient-level longitudinal tracking, time-to-event analysis, or identification of specific treatment sequences — that require individual-level records.
Third, synthetic and aggregated real-world datasets: generated from real-world claims using generative methods with differential privacy guarantees, synthetic health data preserves the statistical properties of the original population without containing records that map to real individuals. Properly constructed synthetic data is GDPR-compliant by design — it is not personal data under the legal definition — which means it can be transferred, analysed, and shared without the data residency constraints that apply to patient-level records. Synthetic data has genuine limitations: it is validated against the source population but may not capture rare events or long-tail distributions with precision, and HTA bodies in different markets have different familiarity with synthetic evidence. But for certain use cases — algorithm development, scenario modelling, early market sizing, training analytical models before real-world study execution — it is not a compromise. It is the right tool.
What GDPR actually demands from a pharma company engaged in European evidence generation is a documented, auditable data processing framework: clear legal basis for processing under Article 9(2), appropriate data processing agreements with vendors and data custodians, anonymisation methodology that meets EDPB guidance on what constitutes genuine de-identification (not just pseudonymisation), and — for any cross-border transfer — appropriate safeguards under Articles 44–49. That documentation is not optional. EDPB guidance on what constitutes valid data processing consent and adequate anonymisation has tightened over the past three years, and national data protection authorities across Continental Europe are actively supervising health data processing in the research context.
The incoming European Health Data Space creates a relevant new dynamic here. EHDS secondary use provisions — which will allow researchers to access health data across EEA member states through a standardised access mechanism — are being phased in between 2025 and 2028 under the EHDS regulation. For companies launching imminently, EHDS is not yet a practical alternative to current national access routes. But for evidence planning that extends into 2026 and beyond, understanding how EHDS will change the data access architecture — and what evidence generated now will need to be supplemented or replicated under EHDS frameworks — is genuinely strategic. The teams currently waiting for EHDS to solve the fragmentation problem before investing in European data capability are, in most cases, going to miss the evidence window for their launch.
Building a Pan-European RWE Strategy Around Your Launch Sequence
This is where the structural understanding becomes commercial strategy — and where the difference between companies that launch successfully and companies that lose reimbursement negotiations becomes visible in retrospect.
The core argument is this: you cannot build a single pan-European RWE study and expect it to serve your evidence needs across multiple national HTA submissions. The heterogeneity of European data systems, combined with the national specificity of payer evidence requirements, means that “one study to rule them all” almost never works in practice. What works is a sequenced, market-tier-aware evidence plan that aligns data source selection with market access priority and launch timing — built early enough that the access processes have time to run.
Start with your market access priority. For most launches, Germany and France are the critical early markets — Germany because it is the largest European market by volume and because national HTA evaluation creates a price reference that affects negotiations elsewhere; France because SNDS-based evidence is essentially required for HAS dossier submissions and because the reimbursement negotiation process is evidence-intensive. Both markets require country-specific data. A study conducted on French SNDS data will not substitute for German-specific claims evidence in a G-BA evaluation, and vice versa. If you are planning a sequential launch with Germany and France as the first two markets, you need country-specific data strategies running in parallel from an early stage — not sequentially after the first market is approved.
Map your evidence questions to data fit-for-purpose, not to vendor coverage. The question to ask for each piece of evidence is not “which vendor covers this country” but “what data type is actually required for this endpoint.” Disease burden and epidemiology can often be addressed with aggregated population-level data or epidemiology datasets — patient-level claims are not always necessary and may not add precision for a prevalence or incidence estimate. Treatment pattern analysis typically requires longitudinal patient-level data with prescription and diagnosis linkage — which means you need country-specific claims access. Outcomes analysis may require linkage to hospital records, cause of death data, or registry data that sits outside the primary claims system. A robust real-world evidence strategy maps each evidence requirement to the data type that is genuinely fit for purpose — not to the data that is most available or most familiar.
Build your evidence plan around access timelines, not study timelines. This is the structural shift that most pharma teams make too late. Regulatory approval processes for national health data — particularly Tier 1 sources like SNDS in France or statutory insurance claims in Germany — can take six to eighteen months from application to data access, depending on the complexity of the study design and the national approval pathway. This timeline is fixed. It cannot be accelerated by commercial urgency or by choosing a vendor with an existing data relationship, because the approval sits with the national data governance authority, not with the vendor. If your launch target is 24 months from now and you need SNDS-based evidence for your HAS dossier, the data access application should be in process now — not at the point of clinical submission.
Use smaller markets strategically. Nordic data — particularly Swedish and Danish registry data — is analytically powerful and often faster to access than the major Continental European sources. For companies that need a rapid, methodologically rigorous outcomes study to anchor their evidence package while larger national data access processes run, Nordic data is frequently the right choice. The evidence generated will not substitute for a German or French country-specific study in those payers’ HTA evaluations — but it demonstrates rigorous methodology, supports publication strategy, and provides a credible reference point for markets that are willing to extrapolate from Nordic populations. Knowing how to use smaller markets’ data to fill evidence gaps while primary market data access matures is a genuine strategic advantage.
The market access evidence requirements for different HTA bodies are genuinely different, and understanding those differences before designing the evidence plan — rather than after — is the skill that most teams lack internally. ISPOR’s guidelines on real-world evidence in decision-making provide the methodological framework; understanding how each national HTA body interprets and applies those guidelines in practice requires market-specific expertise that does not exist in published guidance.
How to Evaluate European RWD Providers Without a Vendor Sales Pitch
The challenge many pharmaceutical companies face in the European data market is not finding providers — it is knowing which questions to ask and how to interpret the answers.
Every major European RWD provider will tell you they have pan-European coverage. The question is what that coverage actually means for your specific evidence question, your specific therapeutic area, and your specific launch timeline. Coverage claims that sound equivalent at the sales stage can diverge dramatically when you test them against a feasibility assessment for a narrow indication in a specific patient population.
When evaluating European-specific data providers, the dimensions that matter — and that most teams underweight — are these.
Coverage model, not just coverage claim. Is the provider’s “pan-European” data based on direct data partnerships with national custodians, or is it aggregated from secondary sources? For statutory insurance claims data, the answer to this question determines whether patient-level longitudinal analysis is actually possible, what the data latency looks like (how recent are the records), and whether the coverage reflects the full insured population or a selected panel. Some providers offer genuine national-level coverage with data residency compliant access; others offer aggregated snapshots that are useful for some purposes and not others. The distinction is not always clear from vendor materials.
GDPR compliance posture, not just GDPR certification. Does the provider have documented anonymisation methodology that is auditable and consistent with EDPB guidance? What is the legal basis they use for processing under Article 9(2), and does it align with the basis you will need to reference in your own data processing framework? Is data processing conducted within the EEA, or are there cross-border transfer mechanisms involved? These are not box-ticking questions — the answers directly affect your GDPR exposure and the validity of your data processing documentation.
Analytical capability and data model transparency. What data model does the provider use? Is it OMOP-compliant, which would enable interoperability with OHDSI network studies? Is the schema documented in sufficient detail to evaluate fit-for-purpose for your specific endpoints? How are ATC codes mapped to specific product-level identifiers, and how is that mapping maintained as formularies change? These are questions that a data science team can evaluate — but only if the provider is willing to be transparent about methodology, which not all are.
Access timeline realism. If a provider is offering access to national statutory insurance data in under three months, it is worth understanding exactly how they are achieving that. Legitimate national data access — particularly for sources like SNDS or German GKV claims — involves regulatory approval timelines that no vendor can compress unilaterally. Access that appears faster may be using a pre-approved study template that limits analytical flexibility, or it may be a different data product than the nationally representative source it is being presented as.
Understanding which data sources match specific use cases requires deep market knowledge that most companies lack internally — and that cannot be reliably acquired from vendor sales processes alone. Expert advisors who understand both the evidence requirements of your specific therapeutic area and the actual capabilities and limitations of the European data vendor landscape can accelerate your path to value while significantly reducing the risk of a costly feasibility failure.
Common Questions from HEOR and Market Access Teams
Can we use one data study across multiple European HTA submissions? Rarely without market-specific supplementation. France’s HAS and Germany’s national HTA framework both have strong preferences for country-specific evidence, and the coding systems, healthcare utilisation patterns, and treatment pathways that a multi-country pooled study captures may not be representative of any individual national market. Cross-market evidence can support secondary analyses and market access arguments in smaller markets — but for Germany and France, expect to need country-specific data.
How long does it realistically take to get access to SNDS data in France? For a standard research application through the Health Data Hub, allow six to twelve months from application to data access for a well-prepared submission. Complex study designs or new applicant organisations with limited track record may take longer. The Health Data Hub’s application guidance is public; the realistic timeline is something HEOR teams consistently underestimate relative to what the published process suggests.
Is synthetic European health data accepted by HTA bodies? For certain use cases — scenario modelling, algorithm validation, health economic modelling inputs — yes, and increasingly so. EMA’s evolving guidance on novel methodologies reflects growing institutional comfort with synthetic and simulation-based evidence where appropriate. For primary effectiveness or safety endpoints in HTA dossiers, patient-level real-world data remains the standard; synthetic data is complementary rather than substitutive in most submission contexts.
What does EHDS actually mean for our evidence planning timeline? EHDS secondary use provisions — allowing cross-border health data access for research — will phase in between 2025 and 2028. Practical implementation across all member states will likely take longer. For launches in 2025–2027, EHDS should be monitored but should not substitute for existing national data access strategies. For portfolio planning beyond 2028, EHDS changes the landscape materially — particularly for companies that need pan-European longitudinal evidence across multiple markets simultaneously.
We’re a small biotech with no internal HEOR function. Where do we start? Start with the evidence requirements for your priority market’s HTA process — understand what that specific payer will require before selecting a data strategy. Then work backwards from the data access timeline to determine when evidence planning needs to begin relative to your development and submission milestones. The mistake most lean organisations make is treating European data strategy as a late-stage procurement decision rather than a clinical development planning input. By the time regulatory approval is visible on the horizon, the data access window for launch-aligned evidence has often already closed.
Your Next Steps with MEDDDICAL
European real-world data fragmentation is not a problem to be solved. It is a landscape to be navigated — strategically, early, and with an accurate map of what actually exists in each market rather than what vendor decks claim exists.
The companies that navigate this landscape most successfully share one characteristic: they begin evidence planning substantially earlier than their peers, and they design that planning around the structural realities of European data access rather than against them. The teams that lose reimbursement negotiations — or that concede price in markets where stronger evidence would have supported a different outcome — almost always do so because the evidence strategy was built too late, or built around data that wasn’t actually fit for the purpose it was being used for.
If you are planning a multi-market European launch and are in the early stages of scoping your evidence strategy, the most valuable investment you can make right now is understanding which data sources are genuinely available for your indication, in your priority markets, within your access timeline. That requires market-specific knowledge that goes beyond what published guidance and vendor sales processes provide.
Contact us to discuss your specific evidence requirements. Whether you’re at the early planning stage, evaluating data providers, or trying to understand why a current evidence strategy isn’t working as expected — a focused consultation with MEDDDICAL’s advisory team can significantly accelerate your path to a defensible, payer-aligned evidence package.
Reach out to explore how expert RWE advisory can accelerate your path to value across Continental European markets.
Sources & References
- European Parliament and Council. Regulation (EU) 2016/679 (GDPR). Official Journal of the European Union. Available at: https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX%3A32016R0679
- European Data Protection Board. Guidelines and Recommendations on Health Data Processing. EDPB Official Documents. Available at: https://edpb.europa.eu/our-work-tools/our-documents_en
- European Commission. European Health Data Space — Regulation and Implementation. Available at: https://health.ec.europa.eu/ehealth-digital-health-and-care/european-health-data-space_en
- European Medicines Agency. DARWIN EU — Data Analysis and Real World Interrogation Network. EMA. Available at: https://www.ema.europa.eu/en/about-us/networks/darwin-eu
- European Medicines Agency. Qualification of Novel Methodologies for Medicine Development. EMA Scientific Advice. Available at: https://www.ema.europa.eu/en/about-us/…
- ISPOR. Real-World Evidence in Health Care Decision Making. ISPOR Resources. Available at: https://www.ispor.org/heor-resources/more-heor-resources/real-world-evidence
- Observational Health Data Sciences and Informatics (OHDSI). OMOP Common Data Model. OHDSI Network. Available at: https://www.ohdsi.org/
- EUnetHTA. Joint Clinical Assessments and HTA Collaboration Framework. Available at: https://www.eunethta.eu/
- Health Data Hub (France). Accessing the SNDS — Application Guidance for Research Projects. Available at: https://www.health-data-hub.fr/en
- Haute Autorité de Santé (HAS). Methodological Guide — Real-World Studies. HAS Publications. Available at: https://www.has-sante.fr/
- Schneeweiss S, et al. Real-world Evidence in the Post-COVID Era. Value in Health. 2021;24(9):1219–1223. Available at: https://www.valueinhealthjournal.com/
- EFPIA. Unlocking the Value of Health Data — The EFPIA Position on the European Health Data Space. Available at: https://www.efpia.eu/