The Importance of Good Clinical Practice in Real-World Evidence Studies

Real-world evidence (RWE) studies represent a significant shift from traditional clinical trials, using data collected outside controlled research environments to evaluate medical products. A 2025 FDA analysis found that regulatory submissions incorporating real-world data (RWD) sources now account for approximately 30% of new drug applications, yet many sponsors struggle to maintain good clinical practice (GCP) standards when working with these diverse data sources.

The challenge is substantial: while traditional clinical trials follow established protocols in controlled settings, RWE studies must ensure data quality and regulatory compliance across electronic health records, insurance claims databases, patient registries, and digital health technologies. The ICH E6(R3) Good Clinical Practice guidance, finalized in January 2025, provides the framework for maintaining scientific rigor in these complex data environments.

Understanding how GCP principles apply to RWE studies isn’t just about regulatory compliance—it’s about ensuring that evidence generated from real-world settings meets the same standards of reliability and validity that regulators expect from traditional trials.

Understanding Real-World Data and Evidence in the GCP Context

Defining Real-World Data Sources

Real-world data encompasses information collected during routine healthcare delivery rather than in controlled research settings. According to FDA guidance, RWD includes data from electronic health records, medical claims systems, product registries, and digital health technologies that capture patient health status outside traditional clinical trials.

The distinction matters for GCP compliance because each data source presents unique quality challenges. Electronic health record data may contain incomplete entries or inconsistent coding practices across healthcare systems. Claims data captures billing information rather than clinical details, potentially missing critical safety events that weren’t reimbursed.

Patient registries offer more structured data collection but vary widely in their data governance standards. Some registries maintain rigorous quality control processes similar to clinical trials, while others rely on voluntary reporting with minimal verification procedures.

Converting Data into Evidence

Real-world evidence represents the clinical conclusions drawn from analyzing RWD to evaluate medical product benefits or risks. The transformation from raw data to regulatory-quality evidence requires adherence to GCP principles throughout the data curation, analysis, and interpretation process.

This conversion process involves three critical phases where GCP standards apply:

• Data extraction and cleaning - ensuring source data accuracy and completeness
• Statistical analysis planning - pre-specifying analytical approaches to avoid bias
• Results interpretation - maintaining objectivity in clinical conclusions

Each phase requires the same level of scientific rigor and documentation that traditional clinical trials demand, despite working with observational rather than experimental data.

Regulatory Framework for RWE Studies

The 21st Century Cures Act of 2016 expanded FDA’s authority to consider RWE for drug approvals, creating new pathways for using observational data in regulatory submissions. FDA’s 2018 RWE Framework established that real-world studies must meet the same evidence standards as randomized controlled trials.

Recent regulatory developments have clarified GCP expectations for RWE studies. The ICH E6(R3) guidance specifically addresses decentralized and pragmatic trial elements, while FDA’s draft Annex 2 guidance provides detailed considerations for applying GCP principles to real-world data sources.

GCP Principles Applied to Real-World Evidence Studies

Data Integrity Requirements

ALCOA++ principles form the foundation of data integrity in RWE studies, just as they do in traditional trials. However, applying these standards to real-world data sources requires different approaches than controlled clinical environments.

Attributable data in RWE studies must trace to specific healthcare encounters and providers, even when working with de-identified datasets. This requires maintaining audit trails that connect analytical datasets back to original source systems without compromising patient privacy.

Legible and Contemporary standards present unique challenges when working with historical healthcare data. Electronic health records may contain abbreviated notes or coded entries that require clinical interpretation, while claims data reflects administrative coding practices that may not align with clinical terminology.

Source Data Verification Approaches

Traditional source data verification (SDV) processes don’t translate directly to RWE studies using large electronic datasets. FDA has adapted its inspection approaches to focus on three key areas for RWD-based submissions:

Data quality assessment evaluates the completeness, accuracy, and consistency of source datasets. Inspectors review data governance procedures, validation processes, and quality metrics used by data custodians such as healthcare systems or registry operators.

Data curation and transformation verification examines how sponsors convert raw healthcare data into analysis-ready datasets. This includes reviewing inclusion/exclusion criteria application, data standardization procedures, and handling of missing or inconsistent values.

Analysis methodology review focuses on pre-specified analytical plans and their execution, similar to traditional clinical trials but with additional attention to confounding factors and selection bias inherent in observational data.

Protocol Development for RWE Studies

RWE study protocols must address the same fundamental elements as traditional clinical trial protocols while accounting for the observational nature of real-world data collection. The study objective should clearly specify the clinical question and target population, including criteria for identifying eligible patients within source datasets.

Primary and secondary endpoints require careful definition when working with real-world data sources. Unlike controlled trials where endpoints are prospectively captured using standardized instruments, RWE studies must rely on available healthcare data elements that may serve as proxies for clinical outcomes of interest.

Statistical analysis plans become particularly critical in RWE studies due to the potential for multiple analytical approaches and post-hoc modifications. Pre-specifying analytical methods, including approaches for handling missing data and controlling for confounding factors, helps maintain scientific objectivity and regulatory acceptability.

FDA Inspection Approaches for Real-World Evidence Studies

Evolution of GCP Compliance Evaluation

FDA’s Office of Scientific Investigations has adapted its inspection methodologies to address the unique characteristics of RWE studies. Traditional site-based inspections remain important, but regulators now employ remote regulatory assessments (RRAs) and alternative oversight approaches for studies incorporating real-world data sources.

Remote regulatory assessments allow FDA to evaluate GCP compliance without physical site visits, particularly valuable when reviewing electronic data systems and centralized analytical processes. These voluntary assessments focus on data governance procedures, quality control measures, and analytical methodology rather than traditional site infrastructure.

The shift reflects practical realities of RWE studies, where “investigational sites” may be healthcare systems contributing data rather than research facilities enrolling patients. RRAs enable more efficient oversight of multi-site studies using electronic health record data from dozens or hundreds of healthcare providers.

Focus Areas for RWE Inspections

FDA inspections of RWE studies concentrate on areas most likely to impact data reliability and regulatory decision-making. Data source qualification receives significant attention, including evaluation of how sponsors assessed the suitability of healthcare databases or registries for their specific research questions.

Data extraction and processing procedures undergo detailed review to ensure that transformation from source systems to analytical datasets maintains data integrity. Inspectors examine database queries, data standardization procedures, and quality control measures applied during dataset creation.

Analytical conduct verification focuses on confirming that final analyses followed pre-specified protocols and that any deviations were appropriately documented and justified. This includes reviewing analysis code, output files, and documentation of analytical decisions made during study execution.

International Collaboration in RWE Oversight

FDA collaborates with international regulatory partners to share information and coordinate oversight of multinational RWE studies. The agency’s partnerships with European Medicines Agency (EMA), UK’s Medicines and Healthcare products Regulatory Agency (MHRA), and Health Canada include specific provisions for sharing inspection findings related to real-world data sources.

These collaborations become particularly valuable for RWE studies that span multiple countries and healthcare systems. Shared oversight reduces duplicative inspections while ensuring consistent application of GCP standards across different regulatory jurisdictions.

Implementation Challenges and Practical Solutions

Data Access and Privacy Considerations

Patient privacy protection requires careful balance with regulatory transparency requirements in RWE studies. Healthcare data custodians often restrict access to protect patient confidentiality, creating challenges for traditional FDA inspection procedures that expect direct access to source documents.

Practical solutions include establishing data access agreements that provide FDA with appropriate review capabilities while maintaining privacy protections. These agreements may specify procedures for remote data review, use of statistical disclosure controls, or provision of data samples that enable regulatory assessment without full dataset access.

Data de-identification procedures must meet both HIPAA requirements and FDA expectations for data traceability. Sponsors need robust procedures for linking de-identified analytical datasets back to source records while maintaining appropriate privacy safeguards.

Quality Assurance in Decentralized Data Environments

Multi-site data harmonization presents significant quality challenges when combining real-world data from diverse healthcare systems. Electronic health record systems use different data models, coding practices, and quality control procedures, requiring careful standardization to create analytically useful datasets.

Effective approaches include establishing data quality metrics that can be applied consistently across participating sites. These metrics should address completeness rates, coding consistency, and temporal data availability relevant to study endpoints.

Vendor oversight becomes critical when working with third-party data aggregators or technology platforms. Sponsors must maintain appropriate oversight of contracted organizations handling data extraction, processing, or analysis functions, ensuring these activities meet GCP standards.

Technology Integration and Validation

Digital health technologies used in RWE studies require validation procedures similar to those used for traditional clinical trial technologies. However, the validation approach may differ when technologies capture real-world usage rather than controlled research environments.

FDA guidance clarifies that sponsors don’t need to inspect individual digital health devices when data transfers to secure electronic repositories according to pre-specified procedures. The first durable database receiving DHT data is considered the source for regulatory purposes.

Electronic data capture systems in RWE studies may differ significantly from traditional clinical trial platforms. These systems must maintain audit trails, user access controls, and change documentation that meet regulatory standards while accommodating the higher data volumes typical of real-world datasets.

Future Directions and Regulatory Evolution

Emerging Guidance and Standards

The ICH E6(R3) Annex 2 draft guidance, released in December 2024, provides the most comprehensive framework to date for applying GCP principles to innovative trial designs incorporating real-world elements. This guidance addresses pragmatic trials, decentralized studies, and hybrid designs that combine traditional clinical trial elements with real-world data sources.

Future regulatory developments will likely address specific technical standards for common RWD sources. FDA’s collaboration with international partners suggests that harmonized approaches to RWE oversight will continue evolving, potentially reducing regulatory burden for sponsors conducting global studies.

Artificial intelligence and machine learning applications in RWE studies represent an emerging area requiring GCP consideration. These technologies may enhance data quality assessment and analytical capabilities while introducing new validation and transparency requirements.

Industry Best Practices Development

Professional organizations are developing specific guidance for applying GCP principles to RWE studies. The Society for Clinical Data Management, International Society for Pharmacoeconomics and Outcomes Research, and other groups are creating practical resources to help sponsors navigate regulatory expectations.

Industry consortiums focusing on RWE methodology are establishing common standards for data quality, analytical practices, and regulatory submission strategies. These collaborative efforts help establish consistent approaches across pharmaceutical companies and reduce regulatory uncertainty.

Academic partnerships between industry and research institutions are advancing methodological approaches for RWE studies while maintaining appropriate regulatory oversight. These collaborations often serve as testing grounds for innovative approaches that may influence future regulatory guidance.

Conclusion

Good clinical practice principles remain fundamental to generating reliable evidence from real-world data sources, even as the regulatory landscape adapts to accommodate innovative study designs. The successful integration of RWE into drug development requires maintaining the same scientific rigor and data integrity standards that have historically supported traditional clinical trials.

FDA’s evolving oversight approaches, including remote regulatory assessments and international collaboration, demonstrate the agency’s commitment to enabling innovation while preserving regulatory standards. Sponsors who proactively address GCP requirements in their RWE study designs will be better positioned to generate evidence that meets regulatory expectations and supports approval decisions.

The future of clinical evidence generation increasingly depends on our ability to maintain scientific standards across diverse data sources and study designs. By applying established GCP principles to real-world evidence studies, the pharmaceutical industry can unlock the potential of observational data while preserving the integrity that regulators and patients rightfully expect.

Sources

1. FDA Real-World Evidence - FDA’s comprehensive overview of real-world data and evidence regulatory framework
2. ICH E6(R3) Good Clinical Practice Guidance - Latest international standards for clinical trial conduct and oversight
3. FDA’s Alternative Approaches to GCP Compliance - FDA training materials on remote regulatory assessments and innovative oversight methods
4. FDA Inspections of Real-World Data Submissions - Analysis of FDA inspection approaches for studies incorporating real-world evidence
5. ICH E6(R3) Annex 2 Draft Guidance - Draft guidance for applying GCP to decentralized and pragmatic trial elements