In December 2025, the FDA announced it was eliminating a major barrier to using real-world evidence in drug and device application reviews. It was one of the most significant regulatory developments in years. And most pharma regulatory affairs teams missed the strategic implications entirely.
That’s not a criticism — it’s an observation about how fast the regulatory landscape for real-world evidence FDA submissions has moved relative to most organisations’ internal understanding of what is now permissible. The FDA’s RWE programme has been quietly expanding its scope since the 21st Century Cures Act in 2016. The EMA’s parallel evolution through the DARWIN EU framework has been similarly progressive. Together, they’ve created a regulatory environment in which real-world data can now legitimately support label expansions, externally controlled pivotal trials, post-market commitments, and — in specific rare disease contexts — primary effectiveness evidence.
Most internal regulatory teams are still operating two or three guidance cycles behind where the agencies actually are.
This post exists to close that gap. Not with a literature summary — you don’t need another overview of why RWE matters. What you need is a clear, current, practically oriented guide to what FDA and EMA now accept, under what conditions they accept it, what methodological standards they apply, and how to structure a submission that succeeds. That’s what follows.
The Evolving Regulatory Framework: FDA and EMA in 2026
Understanding where the agencies are requires understanding how they got there. Both FDA and EMA have evolved their positions on real-world evidence drug approval through a series of guidance documents — each one progressively clarifying and expanding what is permissible. The problem is that most sponsors are benchmarking their submissions against early guidance rather than the current framework.
The FDA Framework
The FDA’s 2018 Framework for the Real-World Evidence Programme was the foundational document — it established the statutory basis under the 21st Century Cures Act and set out the FDA’s intent to evaluate RWE for regulatory decision-making. What followed was a systematic programme of guidance development that most sponsors have not tracked comprehensively.
The 2021 guidance on considerations for RWD and RWE established the foundational questions regulators ask about any real-world data source: is the data relevant to the regulatory question? Is it reliable enough to support regulatory decision-making? These two questions — relevance and reliability — are the analytical framework reviewers apply in practice. Everything else is elaboration.
The 2024 guidance on assessing electronic health records and claims data went considerably further — providing specific criteria for evaluating data fitness for purpose, data quality documentation requirements, and the validation evidence FDA expects sponsors to generate before asserting that a data source is adequate. This is the guidance most sponsors are underutilising.
The December 2025 announcement built on all of this by eliminating requirements that had effectively mandated patient-level data access as a prerequisite for certain regulatory submissions — a significant practical barrier, particularly for sponsors working with aggregated or anonymised data sources. Combined with the 2025 guidance on non-interventional studies, the current framework is substantially more permissive than the framework against which most organisations’ RWE regulatory strategies were designed.
The Advancing Real-World Evidence Programme is worth specific attention here. The FDA’s review of its own experience with RWE submissions — published in peer-reviewed form by Smith et al. (2025) and Innes et al. (2025) — provides the clearest available window into what has worked in practice, what reviewers have found credible, and what has generated the most regulatory friction. These papers are required reading for anyone structuring an RWE FDA regulatory submission in 2026.
The EMA Framework
The EMA’s evolution has been parallel but distinct — and the differences matter enormously for sponsors pursuing simultaneous global submissions.
The 2025 EMA reflection paper on RWD in non-interventional studies is the current definitive EMA position — and it’s meaningfully different from the FDA’s approach in several respects, particularly around protocol pre-registration expectations, confounding adjustment requirements, and data source documentation standards. It cannot be assumed that a study designed to meet FDA expectations will automatically satisfy EMA reviewers.
The DARWIN EU network represents the EMA’s own RWE infrastructure — a federated network of health databases across EU member states that the agency uses for regulator-led studies. Understanding what DARWIN EU can and cannot do is relevant for sponsors whose data strategy intersects with EMA-initiated evidence generation. The third report on regulator-led studies is the most current account of the network’s practical capabilities.
The EU Health Data Space — currently progressing through implementation — will eventually reshape the data infrastructure available for Pan-European RWE regulatory submissions, though its practical implications for the next three to five years are less dramatic than the policy discussion sometimes implies.
The broader convergence between EMA regulatory requirements and the requirements of national HTA bodies in France, the Netherlands, and Scandinavia is a strategic consideration that deserves its own post — but the short version is that the evidence package designed for an EMA submission is increasingly also the evidence package that HTA bodies will evaluate. Designing for one and retrofitting for the other is a more expensive approach than designing for both from the outset.
What FDA and EMA Actually Accept: A Practical Taxonomy
Here’s where the practical value is — a clear breakdown of what each regulatory pathway actually looks like and what the boundaries of acceptance currently are. Alipour-Haris et al. (2024) provide a useful landscape review of RWE regulatory use cases, but what follows goes further into the operational specifics.
Supplemental Indications and Label Expansions
This is the most established pathway and the one with the clearest regulatory precedent. The FDA has accepted RWE as primary or supporting evidence for supplemental NDAs and BLAs across oncology, rare disease, and paediatric indications — a track record that is now substantial enough to provide meaningful guidance on what makes submissions in this category succeed.
What regulators are looking for in label expansion submissions is straightforward to state and harder to execute: a pre-specified analytical plan that was locked before data access, an outcome definition that mirrors what a randomised trial would have measured, a comparator that represents the relevant clinical counterfactual, and a confounding adjustment approach that is methodologically defensible and adequately documented. The Berger et al. (2017) framework for regulatory use of RWE remains a useful reference for structuring the scientific rationale.
The honest reality is that the methodological bar for label expansion submissions has risen materially since the early precedent cases. Studies that were accepted five years ago as adequately designed would face more intensive scrutiny today. Sponsors using historical precedents as their benchmark are likely underestimating current expectations.
External Control Arms and Externally Controlled Trials
The fastest-growing and most misunderstood regulatory use case. The FDA’s 2023 guidance on externally controlled trials is the governing document — and it is substantially more demanding than many sponsors appreciate when they first decide to pursue this design.
The distinction regulators draw — and it’s a critical one — is between an external control used as a formal comparator in a primary efficacy analysis versus real-world data used as a contextual benchmark against which single-arm trial results are interpreted. The methodological bar for the former is considerably higher. When RWE forms the control arm of a pivotal submission, the pre-specification, documentation, and bias assessment requirements approach the standards applied to interventional controls. This is not a shortcut to a simpler trial — it’s a methodologically demanding alternative that is appropriate in specific contexts.
Those contexts are: unmet medical need in settings where randomisation is ethically or practically constrained, rare diseases with limited patient populations, paediatric indications, and situations where historical trial data provides an inadequate comparator for modern clinical practice. Outside these contexts, an external control arm design will face significant regulatory scepticism regardless of how well it is executed.
For deeper technical guidance on synthetic control arm construction methodology, MEDDDICAL’s post on synthetic patient cohorts for clinical trial design covers the construction approach in detail. Hey and Kimmelman (2021) provide a thorough analysis of how FDA has used external controls in regulatory decision-making in practice — essential reading for anyone designing this type of study. The more recent Selukar et al. (2025) analysis of synthetic control arms in haematological malignancies provides current context on where the regulatory frontier is in one of the most active therapeutic areas.
Post-Market Commitments and Post-Approval Requirements
Accelerated approval and conditional marketing authorisation both generate post-market evidence obligations. RWE is increasingly used to fulfil these — but the assumption that post-market studies face lower regulatory scrutiny than pre-approval submissions is one that gets sponsors into serious trouble.
The FDA’s accelerated approval programme has become significantly more rigorous about post-market commitment timelines and methodology following congressional pressure and high-profile cases where confirmatory evidence was delayed. Sponsors who negotiated post-market commitments assuming they would have flexibility on timeline and study design are discovering that flexibility has narrowed considerably.
The strategic opportunity here is to design post-market studies that simultaneously fulfil regulatory evidence obligations and generate the health economic evidence that national HTA bodies across Continental European markets require for reimbursement decisions. The MEDDDICAL post on HEOR simulation and cost modelling covers how aggregated claims data supports both sides of this equation.
Safety Surveillance and Pharmacovigilance
The most established and least contested regulatory use of RWE. Claims data and administrative health data are widely used for signal detection and population-level safety surveillance — and regulatory expectations in this context are well-established and reasonably consistent between FDA and EMA.
What has changed is the data quality standard being applied. Regulators are increasingly specific about coding consistency requirements, minimum follow-up periods, and the documentation needed to support pharmacovigilance-grade analyses. The implicit assumption that pharmacovigilance RWE is held to a lower methodological standard than effectiveness RWE is not accurate — the standard is different, but it is not lower.
Primary Evidence in Rare Disease Contexts
The most emerging and most misunderstood regulatory pathway — and the one where the gap between what is theoretically possible and what is practically achievable is widest.
FDA and EMA will accept RWE as primary effectiveness evidence in rare disease contexts under specific conditions: a patient population too small for a conventionally powered randomised trial, an unmet medical need that makes placebo-controlled randomisation ethically unacceptable, and a natural history that is sufficiently well-characterised to serve as a credible historical control. Wu et al. (2020) remain the definitive reference on the US experience with RWE in rare disease regulatory submissions.
The foundational step is characterising the patient population through rigorous diagnosed prevalence data — the natural history study is only credible if the population it describes is demonstrably representative of the population the drug will treat. MEDDDICAL’s work on diagnosed prevalence and patient data solutions covers this foundational step in detail.
The honest professional view: this pathway is real, it has produced approvals, and the conditions under which it applies are specific enough that most assets won’t qualify. If you think your asset qualifies, the next step is regulatory engagement — not study initiation.
The Methodological Standards Regulators Apply
Understanding what regulators will accept is half the challenge. Understanding how they evaluate it is the other half — and the half that most internal teams underestimate.
Study Design and Pre-Specification
The most consequential methodological requirement in the entire RWE regulatory framework is pre-specification. The analytical plan must be locked before data access begins. Not before analysis begins. Before access. This is non-negotiable for any submission where the analytical results will be used as primary or supporting regulatory evidence.
The reason is obvious once stated: an analytical plan developed after observing the data cannot be distinguished from one that was optimised to produce favourable results. Regulators know this, and a submission that cannot demonstrate clear pre-specification faces scepticism that is very difficult to overcome retrospectively.
The target trial emulation framework — formalised by Hernán and Robins (2016) and increasingly endorsed by FDA as the methodological gold standard for observational studies — provides the most rigorous approach to pre-specification. It requires the sponsor to explicitly define the hypothetical randomised trial being emulated: the eligibility criteria, the treatment strategies, the outcome, the follow-up period, and the causal contrast of interest. Every departure from this hypothetical trial design is then a potential source of bias that must be identified, assessed, and addressed. The target trial emulation framework is worth the investment to learn properly. Studies designed with it are substantially more likely to survive regulatory scrutiny than those designed without it.
Protocol registration in ClinicalTrials.gov or the EU Clinical Trials Register is increasingly expected as a transparency measure — and the absence of pre-registration is increasingly used by reviewers as a basis for scepticism about pre-specification claims. Register the protocol. There is no good reason not to.
Data Quality Documentation
The FDA’s 2024 guidance on assessing EHR and claims data provides the most specific available articulation of what regulators expect sponsors to demonstrate about data quality. The core requirements are: evidence that the data source captures the population relevant to the regulatory question, evidence that the coding and data completeness are adequate for the specific outcome and exposure variables being analysed, and external validation of key variables against independent sources where available.
What this means in practice is that data quality documentation needs to be generated prospectively — before the primary analyses are built on top of the data — rather than assembled retrospectively to respond to regulatory questions. Sponsors who treat data quality documentation as a submission requirement rather than a study design requirement consistently encounter problems during review. The ISPOR RWE Transparency Initiative guiding principles and the RECORD reporting standards provide the documentation framework that regulators expect to see applied.
For the broader clinical dataset landscape that underpins regulatory-grade external control designs, MEDDDICAL’s overview of clinical and preclinical datasets powering modern drug development covers the data architecture questions that regulatory submissions depend on.
Bias Assessment and Sensitivity Analyses
Sensitivity analyses are the area where most submissions are least prepared. Regulators will request them — systematically and specifically. Building them into the primary analysis plan rather than generating them in response to regulatory questions saves months from the review timeline and demonstrates the methodological confidence that supports regulatory credibility.
The minimum sensitivity analysis package for a label expansion or external control submission includes: an alternative confounding adjustment approach, analyses in pre-specified subgroups defined by clinical relevance, analyses using alternative outcome definitions, and analyses exploring the impact of missing data assumptions. The GRACE principles provide a useful framework for structuring the bias assessment that should accompany the sensitivity analysis package. Cohort definition quality is equally fundamental: phenotype validation in claims data — including documented sensitivity, specificity, and PPV metrics — is increasingly expected as part of the study protocol package for RWE regulatory submissions.
Propensity score methodology deserves specific mention because it is both widely used and widely misapplied in regulatory submissions. Reviewers are sophisticated about propensity scoring — they know how to assess whether covariate selection was adequate, whether balance was achieved and demonstrated, and whether the method was applied to match populations or to adjust for confounding within a regression framework. Applying propensity scoring without adequate covariate selection or without demonstrating post-matching balance is worse than not applying it — it signals methodological sophistication without demonstrating methodological rigour.
Common Regulatory Submission Mistakes That Get RWE Rejected
These are the ones worth naming directly, because they appear in submissions from organisations that should know better.
Study design defined after data exploration. The most common and most consequential mistake. If the analytical plan was not locked before the data was accessed, the submission is methodologically compromised in a way that is very difficult to recover from. Regulators have become adept at detecting post-hoc analytical decisions. Sherman et al. (2016) articulated the fundamental challenge clearly; the CIOMS Working Group XIII report (2025) provides the most current regulatory perspective on how agencies are approaching this.
Data source selected before the research question. Choosing the data source that is available and then defining the research question it can answer is precisely backwards. The research question defines what data is needed. A submission built on the available dataset rather than the appropriate dataset is answering a different question than the one the regulator needs answered.
Inadequate confounding adjustment. Propensity scoring applied without adequate covariate selection. Regression adjustment without assessment of residual confounding. Target trial emulation claimed but not executed. Regulators ask for the covariate selection rationale. Have it ready.
Missing sensitivity analyses. Submitting primary results without the robustness checks regulators will request adds months to the review timeline. Build sensitivity analyses into the primary analysis plan and submit them with the initial package.
Insufficient data quality documentation. Asserting that the data source is fit for purpose without validation evidence. Regulators cannot accept the assertion — they need the documentation. Kesselheim and Doherty (2021) describe the challenge clearly from a regulatory policy perspective.
Failing to engage regulators before study initiation. This one is both the most preventable and the most consequential. Pre-submission regulatory engagement on study design — before data access, before analytical commitment — is the single most effective risk mitigation available to sponsors pursuing RWE regulatory pathways. It is underutilised because most teams don’t know what to bring to these meetings or what questions to ask. The next section covers this.
Designing for FDA and assuming EMA alignment. The two frameworks are not identical. Sponsors pursuing simultaneous submissions to both agencies need to understand where the requirements converge and where they diverge — and design studies that satisfy both from the outset rather than discovering the gaps during review.
How to Structure a Successful RWE Regulatory Submission
Pre-Submission Regulatory Engagement
This is the single highest-leverage action available to any sponsor planning an RWE regulatory submission. The FDA Type C meeting and the EMA scientific advice procedure both allow sponsors to present a proposed study design and receive regulatory guidance before analytical commitment is made. The value is not just avoiding rejection — it’s understanding specifically what evidence the agency will find compelling for the question being asked.
The FDA’s Model-Informed Drug Development Paired Meeting Programme is specifically relevant for sponsors incorporating quantitative methods — including Bayesian historical borrowing and synthetic control approaches — into their regulatory strategy. Galluppi et al. (2024) provide practical guidance on preparing for these meetings.
What to bring: a clearly articulated research question, a proposed data source with initial fitness-for-purpose documentation, a draft target trial emulation specification, a proposed confounding adjustment approach with covariate selection rationale, and specific questions about where regulatory guidance would be most valuable. What not to bring: a concept without a protocol, or a completed analysis seeking retrospective endorsement.
Study Protocol and Documentation Requirements
A regulatory-grade RWE study protocol is substantially more detailed than a typical observational study protocol submitted for publication. It must include: the precise research question and target trial specification, the data source justification with fitness-for-purpose evidence, the patient population definition including all inclusion and exclusion criteria, the exposure and comparator specification, validated outcome definitions with ascertainment methodology, the covariate selection rationale for confounding adjustment, the complete statistical analysis plan including all pre-specified sensitivity analyses, and the bias assessment framework identifying potential sources of bias and how each will be addressed.
The ISPOR-ISPE good practices guidance and the Schneeweiss et al. (2019) graphical depiction framework are the methodological references regulators will expect to see cited in the protocol rationale.
Data Source Justification
The data source justification section is where submissions most commonly fall short. It is not sufficient to describe the data source and assert its appropriateness. The justification must demonstrate: coverage adequacy for the target population, validation of key outcome and exposure variables, comparison of population characteristics against external benchmarks, and honest disclosure of limitations with an explanation of why those limitations do not compromise the regulatory conclusions.
That last point — honest disclosure — is worth emphasising. Regulators are not looking for data sources without limitations. They are looking for sponsors who understand their data’s limitations, have assessed their impact on the conclusions, and can demonstrate that the conclusions are robust to those limitations. Hiding data quality concerns in a submission is less effective than disclosing them with an accompanying sensitivity analysis that shows the conclusions are robust.
Evidence Package Structure
An RWE submission package for a supplemental NDA or BLA differs from a traditional clinical trial submission primarily in Module 5 — the clinical study reports and tabulated data. The specific modules where RWE evidence is presented depend on the regulatory context: as primary effectiveness evidence, as a historical comparator in an externally controlled trial, or as supportive evidence alongside randomised trial data.
The integrated summary of effectiveness is the document that most requires careful construction when RWE supplements randomised evidence — the two evidence streams need to be integrated coherently, with a clear explanation of why each contributes to the regulatory conclusion and how potential inconsistencies between them are interpreted.
The Strategic Opportunity: Building RWE Regulatory Capability Now
The regulatory acceptance of real-world evidence for drug approval is not static. It’s expanding. And the expansion is not linear — it’s compounding, as each accepted precedent lowers the bar for the next.
Companies that build RWE regulatory capability now — understanding what is accepted today and positioning for what will be accepted as the guidance landscape continues to develop — gain strategic advantages that competitors who treat RWE as a post-approval afterthought cannot easily close. Dagenais et al. (2022) describe how RWE can inform development strategy and clinical trial design well before submission — the regulatory submission is downstream of evidence generation decisions made eighteen to thirty-six months earlier.
The therapeutic areas where RWE regulatory pathways are most established right now — oncology, rare disease, immunology, and neurology — are also the areas where the strategic value of earlier regulatory engagement is highest. If your pipeline is concentrated in any of these areas and your regulatory strategy does not yet incorporate RWE as a primary evidence pathway, the gap between where you are and where your most sophisticated competitors are is probably larger than you think.
The evidence infrastructure decisions that need to be made now — data source partnerships, natural history study design, outcome definition validation — cannot be made at the submission planning stage. They need to be made now, before expensive evidence generation commitments are locked in, to ensure that the right data is available when the submission requires it. The MEDDDICAL RWE strategy and analytics overview covers the broader landscape within which these decisions sit.
Understanding which combination of data coverage, methodological approach, and regulatory documentation actually fits your specific submission context is not something that can be determined from guidance documents alone. The gap between what the guidance documents permit and what a specific reviewer will find credible for a specific indication is where regulatory expertise — and the advisory support that provides it — creates genuine value.
Your RWE Regulatory Submission Checklist: Where to Start
If you’re planning a regulatory submission where RWE could play a role — whether as primary evidence, an external control arm, or a post-market commitment — here is where to start.
- Define the specific regulatory question. Is this a label expansion? An externally controlled pivotal trial? A post-market commitment? A pharmacovigilance study? The regulatory pathway governs everything that follows. Don’t start designing a study before the regulatory question is precisely articulated.
- Identify the governing guidance documents. For each regulatory pathway, there are specific FDA and EMA guidance documents that govern the evidence standard. Know them. The 2023 externally controlled trials guidance, the 2024 EHR and claims data guidance, the 2025 non-interventional studies guidance, and the EMA 2025 reflection paper are the documents that matter most for the contexts discussed in this post.
- Plan pre-submission regulatory engagement before study design is committed. Identify the appropriate meeting type — FDA Type C, MIDD Paired Meeting, EMA scientific advice — and prepare a study design proposal detailed enough to generate specific regulatory guidance. Engage before data access begins.
- Get expert advisory support for data source selection. The regulatory acceptability of a data source depends on coverage, validation, coding quality, and GDPR compliance factors that are not visible from vendor materials alone. Understanding which Pan-European data sources meet the regulatory standard for specific therapeutic areas and submission contexts requires deep market knowledge that most organisations don’t have internally.
- Start earlier than feels necessary. The evidence generation timeline for a regulatory-grade RWE submission — from data source assessment to protocol finalisation to analytical execution to submission — is almost always longer than it looks from the planning stage. The submissions that succeed are the ones that were designed with enough lead time to do everything properly.
The conversation that saves the most time and the most money is the one that happens before any data source is contracted and any study design is committed. That’s the conversation MEDDDICAL is built for. Contact us to discuss your RWE regulatory submission strategy — we’ll help you define what evidence you need, which data sources can generate it, and how to structure the pre-submission engagement that gives your submission the best possible chance of success.
And if you’re coming to this post from the evidence generation side — wondering what analysis-ready datasets look like in practice for a programme like the ones described here — the MEDDDICAL post on analysis-ready real-world evidence and healthcare data for pharma covers the data architecture questions directly.
Sources & References
- FDA (December 2025). FDA Eliminates Major Barrier to Using Real-World Evidence in Drug and Device Application Reviews. Press Announcement.
- Innes et al. (2025). Real-world evidence in new drug and biologics license application approvals during fiscal years 2020–2022. Clinical Pharmacology & Therapeutics.
- Smith et al. (2025). FDA’s Advancing Real-World Evidence Program: initial experience. Clinical Pharmacology & Therapeutics.
- FDA (2018). Framework for FDA’s Real-World Evidence Program.
- FDA (2021). Considerations for the Use of RWD and RWE to Support Regulatory Decision-Making for Drug and Biological Products.
- FDA (2024). Real-World Data: Assessing EHRs and Medical Claims Data to Support Regulatory Decision-Making.
- FDA (2025). RWE: Considerations Regarding Non-Interventional Studies for Drug and Biological Products.
- EMA (2025). Reflection Paper on the Use of Real-World Data in Non-Interventional Studies to Generate Real-World Evidence.
- EMA. Data Analysis and Real-World Interrogation Network (DARWIN EU).
- Prilla et al. (2024). Real-world evidence to support EU regulatory decision making — results from a pilot of regulatory use cases. Clinical Pharmacology & Therapeutics.
- FDA (2023). Considerations for the Design and Conduct of Externally Controlled Trials for Drug and Biological Products.
- Hey & Kimmelman (2021). The use of external controls in FDA regulatory decision making. Clinical Pharmacology & Therapeutics.
- Selukar et al. (2025). Synthetic control arms and other uses of external data in clinical trials for hematological malignancies. Blood Reviews.
- Alipour-Haris et al. (2024). Real-world evidence to support regulatory submissions: a landscape review and assessment of use cases. Clinical and Translational Science.
- Wu et al. (2020). Use of real-world evidence in regulatory decisions for rare diseases in the United States. Pharmacoepidemiology and Drug Safety.
- Hernán & Robins (2016). Using big data to emulate a target trial when a randomized trial is not available. American Journal of Epidemiology.
- ISPOR-ISPE Special Task Force. Good Practices for Real-World Data Studies of Treatment and/or Comparative Effectiveness.
- Benchimol et al. (2015). The RECORD Statement. PLOS Medicine.
- Schneeweiss et al. (2019). Graphical depiction of longitudinal study designs in health care databases. Annals of Internal Medicine.
- Dreyer et al. (2018). GRACE Principles: Recognizing high-quality observational studies of comparative effectiveness. American Journal of Managed Care.
- ISPOR (2022). Real-World Evidence Transparency Initiative: Guiding Principles.
- Kesselheim & Doherty (2021). Real-world evidence in drug regulation — challenges and opportunities. Health Affairs.
- Hennessy et al. (2025). Real-world data and RWE in regulatory decision making: CIOMS Working Group XIII report. Pharmacoepidemiology and Drug Safety.
- Sherman et al. (2016). Real-world evidence — what is it and what can it tell us? New England Journal of Medicine.
- FDA. Model-Informed Drug Development Paired Meeting Program.
- Galluppi et al. (2024). Preparing for the FDA Model-Informed Drug Development Paired Meeting Program. Clinical Pharmacology & Therapeutics.
- Dagenais et al. (2022). Use of real-world evidence to drive drug development strategy and inform clinical trial design. Clinical Pharmacology & Therapeutics.
- European Commission. Proposal for a Regulation on the European Health Data Space.
- FDA. Real-World Evidence Programme — Advancing Real-World Evidence.
