Real-world evidence (RWE) has stepped into that gap. It's not here to replace randomized trials — it complements them, filling in blind spots around feasibility, population diversity, and post-approval performance. This article breaks down what RWE actually is, how it's reshaping trial design decisions, where regulators stand, and what the challenges look like in practice.
Key Takeaways
- Trial design now draws on real-world data (EHRs, claims, registries, wearables)—not only post-market safety monitoring
- FDA and EMA accept RWE for label expansions, external control arms, and select approval decisions
- Eligibility criteria, site selection, and feasibility can be refined with RWE before a protocol is finalized
- Data quality, standardization, and bias remain the top barriers to reliable RWE use
- Multi-regional trials, especially those involving LMICs, gain broader population representation through RWE
What Is Real-World Evidence (RWE) and How Does It Differ From Traditional Trial Data?
The FDA defines real-world data (RWD) as patient health information routinely collected from sources like electronic health records, medical claims, product registries, and patient-generated data such as wearables. Real-world evidence (RWE) is the clinical evidence generated by analyzing that data, as outlined in the FDA's Real-World Evidence Program.
The distinction from a traditional randomized controlled trial (RCT) comes down to purpose and setting:
- RCTs operate in controlled, curated conditions designed to isolate cause and effect
- RWE draws from routine clinical care, showing how treatments perform across broader, more diverse populations
- Development use differs too: RCTs typically support primary efficacy claims, while RWE informs external validity, labeling expansions, and post-market decisions
That messiness cuts both ways. RWE can reveal how a drug behaves in patients who'd never qualify for a tightly controlled trial. But it also brings real limitations: confounding variables, selection bias, and inconsistent outcome capture across health systems. A patient's outcome recorded in one EHR system might be missing entirely in another.
Pragmatic Trials: A Middle Ground
Pragmatic trials sit between these two worlds. They keep some randomization but conduct it in real-world care settings, capturing everyday clinical variability while preserving causal inference. They function as a bridge between explanatory RCTs and observational RWE—useful when you need causal structure without fully leaving routine care.
How Real-World Evidence Shapes Clinical Trial Design and Decisions
RWE's biggest impact happens before a single patient is enrolled. Sponsors use it across five design decisions that determine whether a protocol can recruit, measure, and support regulatory review. Pre-trial design and feasibility. Sponsors mine claims and EHR data to map target populations, stress-test eligibility criteria, and tighten inclusion/exclusion rules before lock. One caution: a 2019 JAMA Network Open study found that only 15% of a studied trial population could be replicated from real-world data alone. RWE tests feasibility; it does not guarantee it. External control arms and hybrid designs. RWE has made its strongest mark here, especially in rare diseases where randomization is ethically or practically difficult. Blinatumomab (Blincyto) for relapsed/refractory leukemia is a clear case: a 189-patient single-arm Phase II study was compared with historical standard-of-care outcomes using propensity-score methods, supporting FDA approval in December 2014. Adaptive trial design. RWE-informed safety signals can trigger protocol amendments or interim go/no-go calls mid-trial, instead of waiting for a fixed analysis point. That shortens the lag between emerging risk and a documented design change. Endpoint selection and sample sizing. Real-world disease-progression data helps sponsors set expected event rates and size cohorts more realistically, reducing underpowered studies and avoidable protocol amendments later. Site selection for multi-regional trials. Recruitment speed and data infrastructure differ sharply across healthcare systems, so site performance in one region rarely transfers unchanged to another. At DRK Research Solutions, our clinical trials implementation team folds local healthcare context into site feasibility for multi-regional protocols so sponsors can prioritize sites that match both the population and the data environment.
Regulatory Perspectives: FDA, EMA, and Global Acceptance of RWE
The FDA's RWE framework traces back to the 21st Century Cures Act of 2016, formalized in its 2018 guidance document. It covers using RWE to support new indications for approved drugs and to satisfy post-approval study requirements.
A 2025 peer-reviewed analysis found RWE in 55 of 218 FDA labeling-expansion approvals (about 25%), holding relatively steady across 2022, 2023, and partial 2024 data. The authors caution against reading this as a hard upward trend given the short observation window, but it confirms RWE has a consistent, if selective, presence in FDA decisions.
The EMA takes a more explicitly case-by-case approach. Its Real-World Data Quality Framework doesn't set fixed acceptance thresholds; assessors judge fitness for purpose individually. DARWIN EU, launched in 2022, standardizes healthcare data across EU databases so regulators can draw timely evidence on medicine safety and effectiveness across a product's lifecycle.
For sponsors running multi-regional programs, acceptance is not uniform:
- FDA use of RWE is selective but documented in roughly one in four labeling expansions
- EMA judges RWE fitness for purpose case by case, without fixed thresholds
- Evidentiary expectations differ by jurisdiction, so a single RWE package rarely satisfies every region
Benefits and Challenges of Using RWE in Clinical Development
RWE can improve trial design and development decisions, but only when teams weigh the gains against data, privacy, and analytical limits.
Benefits sponsors typically see:
- Broader population representativeness than tightly controlled RCT cohorts
- Faster, data-driven feasibility assessment before protocol lock
- Stronger support for underserved and high-risk patients who rarely qualify for traditional trials
Challenges to plan for up front:
- Data quality and standardization gaps. EHR and claims data vary widely in structure across systems. Common data models like OMOP help, but as peer-reviewed research on data quality notes, standardization alone does not guarantee reliability.
- Privacy regulation. HIPAA governs how covered entities can use protected health information for research in the US, often requiring IRB waivers or data-use agreements before cohorts can be assembled.
- Analytical complexity. Separating genuine treatment effect from confounding requires advanced statistical methods, not simple comparisons.
These challenges compound in emerging markets and LMICs, where data infrastructure is often less mature. That reality shapes how DRK approaches vendor selection and data governance across the regions we operate, applying ICH-GCP and GxP standards consistently even where local systems are still developing.
Real-World Examples of RWE in Clinical Development
RWE is already shaping labels, safety monitoring, and trial design. These cases show how sponsors and agencies put it to work in practice.
Label expansion via observational data. FDA review documents cite RWE studies submitted with supplemental applications. Real-world safety and effectiveness data actively inform post-approval label decisions.
Vaccine safety surveillance. The CDC's Vaccine Safety Datalink pulls EHR data from member healthcare sites to detect adverse events in near-real time. FDA COVID-19 vaccine surveillance used this passive monitoring to identify rare risks such as myocarditis, occurring in fewer than 1 in 200,000 vaccinated individuals. A pre-approval trial of typical size would almost certainly have missed that signal.
External comparator arms in oncology. The Blincyto approval shows how historical real-world outcomes can stand in for a randomized control group when enrolling a true control arm is not feasible.
How DRK Research Solutions Integrates RWE Into Trial Design and Delivery
Multi-regional trials succeed or stall on local context. DRK's clinical trials implementation and medical affairs teams factor real-world data insights into protocol design across the healthcare systems we serve in Europe, the Middle East, Asia, Africa, and the Americas.
Our patient-centric, LMIC-focused approach means eligibility criteria and feasibility assessments aren't built on a single region's assumptions. A protocol that works in one healthcare system often needs real adjustment in another, especially where patient populations, care patterns, and data infrastructure differ.
In practice, RWE shapes DRK's design work in several ways:
- Eligibility criteria grounded in real care patterns, not one-market norms
- Feasibility checks against local site capacity and data infrastructure
- Endpoints and visit schedules aligned with how patients actually receive care
Sponsors building an RWE-informed development strategy can work with DRK's clinical development and CRO teams on protocol design, drawing on more than a decade of experience across underserved and emerging markets.
Frequently Asked Questions
What is real-world evidence in clinical trials?
RWE is clinical evidence derived from analyzing real-world data such as EHRs, claims, and registries. It reflects routine clinical care rather than the controlled conditions of a traditional trial.
How does real-world evidence differ from traditional clinical trials?
RCTs use controlled, randomized designs built for causal inference. RWE draws from observational data collected during everyday clinical practice, offering broader but less controlled insight.
What are examples of real-world evidence used in clinical trials?
Common examples include label expansions supported by observational data, external control arms in oncology and rare disease studies, and post-market safety surveillance through registries.
Can real-world data replace randomized clinical trials?
Generally, no. RWD complements RCTs rather than replacing them, and its suitability depends on the research question, data quality, and whether causal inference is required.
How do regulators like the FDA view real-world evidence?
The FDA's RWE framework supports label expansions and post-approval studies when data quality and study design are robust. It has factored into roughly a quarter of recent labeling-expansion approvals.
What are the biggest challenges in using RWE in drug development?
Data quality and standardization gaps, privacy regulations like HIPAA, and the analytical complexity of separating true treatment effects from confounding variables are the core hurdles.
