Real-world data (RWD)

Real-world data (RWD) refers to data relating to patient health status or the delivery of health care that is routinely collected from a variety of sources outside the setting of conventional interventional clinical trials. Main sources are electronic health records, administrative and claims databases (such as France's SNDS), product and disease registries, natural history cohorts, pharmacy and laboratory data, patient-generated data from apps and wearables, and increasingly health data warehouses built by hospitals.

RWD is the raw material from which real-world evidence is derived. It is used across the product lifecycle: to understand disease epidemiology and unmet need, to design and recruit trials, to build external control arms, to support regulatory decisions in rare diseases, to fulfil post-authorisation safety and effectiveness commitments, and to inform health technology assessment and pricing. The EMA's DARWIN EU network, the FDA's RWE framework and the European Health Data Space all aim to make RWD more accessible and reliable. Data quality, completeness and comparability vary widely between sources, and study designs must address confounding and missing data.

RWD is almost always personal data, typically pseudonymised health data processed for secondary use. Under the GDPR this requires a research legal basis (Art. 6(1)(e) or (f) with Art. 9(2)(j)), Art. 89 safeguards, compliance with national access rules (French reference methodologies or CNIL authorisation, Finnish Findata, German Forschungsdatenzentrum), a DPIA and, where the RWD holder is outside the EEA, a transfer mechanism. Contracts with data holders should clarify controller roles and whether the delivered dataset is anonymised or pseudonymised, since this determines the entire compliance burden downstream. See iliomad's health data warehouse services.