On March 16, 2026, the EMA’s Committee for Medicinal Products for Human Use (CHMP) adopted a Data Quality Framework (DQF) for Real-World Data (RWD) to support regulatory use of Real-World Evidence (RWE). The framework extends the existing EMRN DQF and provides structured, actionable guidance for assessing data quality in regulatory submissions and decision-making.
Key Objectives
- Establish a harmonised approach to evaluating RWD quality in regulatory contexts
- Support fit-for-purpose assessment of datasets for specific research questions
- Enable transparent and consistent use of RWE across the product lifecycle
- Align with EU initiatives (e.g., EHDS, DARWIN EU) and existing standards (ICH, ENCePP)
Core Framework Components
The RW-DQF is structured around three main pillars:
- Systems & Process Characterisation
- Assessment of how data are collected, processed, and governed
- Includes maturity model and checklist for data provenance, governance, and infrastructure
- Data Quality Metrics
- Quantitative assessment across key dimensions:
- Reliability (accuracy, traceability)
- Extensiveness (completeness, coverage)
- Coherence (consistency, standardisation)
- Timeliness (data currency)
- Metrics support DQ assessment, assurance, and reporting
- Quantitative assessment across key dimensions:
- Fitness-for-Use Assessment
- Evaluates whether data are suitable for a specific research question
- Requires alignment of dataset with:
- Study design
- Population
- Variables (exposure, outcomes, confounders)
Key Regulatory Principles
- No fixed thresholds: Data quality acceptability is context-dependent
- Question-driven assessment: “Relevance” is defined by the research question
- Lifecycle responsibility: Data quality is shared across stakeholders(data holders, users, regulators)
- Transparency requirement: Detailed documentation of data processes and limitations is expected
Important Considerations for RWD
The framework highlights challenges specific to RWD:
- Heterogeneity of sources, formats, and coding systems
- Secondary use limitations (data not originally collected for research)
- Data linkage and pooling risks (e.g., misclassification, loss of precision)
- Data drift over time affecting reliability
- Privacy constraints limiting access to patient-level data
Implications for Industry
- Sponsors must provide robust data quality documentation in submissions
- Increased emphasis on:
- Metadata transparency
- Traceability and lineage
- Standardisation (e.g., CDMs, vocabularies)
- Early feasibility and relevance assessments will be critical
- May impact study design, data source selection, and regulatory strategy
Regulatory Impact
- Strengthens EMA expectations for RWE use in regulatory decision-making
- Supports broader EU data initiatives and integration of RWD into regulatory frameworks
- Provides a non-prescriptive but structured approach adaptable across use cases
Bottom Line
The EMA RW-DQF represents a major step toward standardising RWD quality assessment, reinforcing that data must be demonstrably fit-for-purpose—not just available—to support regulatory decisions.