EMA publishes final real world data chapter of EU Data Quality Framework

The EMA has released the final real‑world data (RWD) chapter of its Data Quality Framework for EU medicines regulation, further setting out practical recommendations to assess data quality and strengthen the role of real-world evidence (RWE) in EU regulatory assessments.

The European Medicines Agency (EMA) has published the final version of its guidance “Data Quality Framework for EU medicines regulation: application to real-world data (RW-DQF),” marking a further step in supporting the structured use of RWD in regulatory decision-making.

Adopted by the Committee for Medicinal Products for Human Use (CHMP) on March 16, 2026 and released via the EMA website on March 27, the chapter builds on the broader 2023 Data Quality Framework and follows a period of public consultation on a draft version issued in November 2024.

The RW-DQF introduces recommendations intended to support a more consistent and transparent approach to evaluating the quality of RWD used to generate RWE in the assessment of medicines. As regulatory reliance on RWE continues to expand, the framework aims to strengthen confidence in how such data are selected, assessed, and used to inform benefit–risk decisions.

Developed in collaboration with the Heads of Medicines Agencies (HMA) and the Towards European Health Data Space (TEHDAS) initiative, the RW-DQF reflects wider European efforts to enable more effective use of secondary health data. While aligned with developments linked to the European Health Data Space, the document focuses specifically on the challenges associated with using RWD in regulatory contexts, where questions of data provenance, completeness, and reliability remain central.

The guidance extends the European Medicines Regulatory Network (EMRN) Data Quality Framework, providing more targeted recommendations tailored to RWD. It outlines how data quality should be assessed across multiple dimensions, including the characterization of the systems and processes underpinning data and their potential impact on quality. It also introduces metrics to evaluate different aspects of data quality and supports a “fitness-for-use” approach, whereby datasets are assessed in relation to a clearly defined research question. Practical tools, including example checklists, are included to support implementation across different use cases.

EMA positions the RW-DQF as a companion to the EMRN framework and a practical reference for regulatory use. In the document, the Agency notes that “the application of the RW-DQF is strongly encouraged as a best practice framework for guiding the assessment of RWD quality in regulatory contexts.” Rather than introducing prescriptive requirements, the framework supports structured evaluation without defining fixed thresholds or prioritizing specific metrics. Instead, it enables assessors to determine whether a dataset is appropriate based on its intended use and regulatory context, allowing for flexibility across different data sources.

The guidance applies primarily to secondary data generated through routine clinical practice, including electronic health records, claims data, prescription data, and patient registries. It is intended to complement existing methodological standards, such as those developed by the European Network of Centres for Pharmacoepidemiology and Pharmacovigilance (ENCePP) and the International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use (ICH), including the M14 guideline. It should be used alongside these frameworks rather than as a standalone methodological manual.

The RW-DQF is intended for a broad set of stakeholders involved in regulatory evidence generation, including regulators within the EMRN, pharmaceutical companies, contract research organizations, and data networks such as DARWIN EU. It also has relevance for academic researchers, patient organizations, and health technology assessment bodies involved in evaluating evidence derived from RWD.

Looking ahead, the EMA notes that the RW-DQF will continue to evolve alongside advances in data generation and use. Future iterations may expand to cover emerging sources of RWD, including patient-reported outcomes, preference data, wearable devices, mobile health technologies, and social media.