Regulatory decision-making depends on evidence that reflects how medical products perform in real-world use, not only in controlled trials. For more than a decade, the Food and Drug Administration has relied on real-world data and real-world evidence to monitor postmarket safety, inform approvals, and update labeling.1 That direction was reinforced by the 21st Century Cures Act of 2016 and the FDA’s 2018 framework describing how real-world evidence can support new indications and post-approval requirements for drugs and biologics.
That history matters for life sciences teams navigating development, safety surveillance, and ongoing oversight today. As FDA use of real-world evidence continues to expand, expectations around data quality, transparency, and reproducibility are rising. In this blog, we examine how the FDA’s track record already reflects those expectations and what it signals for organizations working to generate regulatory-ready evidence.
Where FDA use of real-world evidence becomes decision-grade
When you look across FDA regulatory decisions that relied on real-world evidence, a pattern emerges. Whether the outcome is a labeling change, a safety communication, or an approval, the evidence that holds up comes from real-world data environments designed for ongoing regulatory use.1
Sentinel as the foundation for real-world safety oversight
That pattern is most visible in the FDA’s Sentinel Initiative, which was created to actively monitor the safety of marketed medical products using routinely collected healthcare data. Several of the agency’s most instructive safety decisions show how this infrastructure shapes regulatory outcomes and what it takes for real-world data to operate at that level.
Sentinel depends on real-world data that can meet demanding operational and regulatory expectations, including the ability to be:
- Standardized into a common data model
- Governed for privacy and appropriate use
- Reanalyzed as new safety questions arise
In 2023, HealthVerity worked directly with the FDA’s Sentinel Innovation Center on a feasibility study to determine whether electronic health record (EHR) data from the vast HealthVerity ecosystem could be converted into the Sentinel Common Data Model and used for active safety surveillance.2 That work focused on data structure, clinical completeness, and whether the data could support repeatable, regulator-grade analyses over time. Preparing data to meet Sentinel expectations is what allows safety monitoring to move beyond claims-only signals and into richer clinical context.
Sentinel studies: when real-world data drives safety decisions
Those requirements are reflected in how Sentinel studies have informed FDA safety decisions.
In one Sentinel analysis, FDA evaluated the risk of hypoglycemia among pediatric patients treated with beta blockers. Using a retrospective cohort design across large, longitudinal datasets, the study identified an association that had not been fully characterized during preapproval development. The findings informed updates to product labeling to better reflect hypoglycemia risk in children and in patients who may not be able to recognize or communicate symptoms. This is a clear example of how Sentinel-grade real-world data can surface clinically meaningful risk in routine care and translate directly into regulatory action.1
Other Sentinel analyses illustrate a more nuanced role for real-world evidence. In an evaluation of interstitial lung disease among patients treated with vedolizumab or natalizumab for inflammatory bowel disease, FDA did not observe an increased risk signal in the Sentinel data. When considered alongside adverse event reports and other sources, however, the agency updated the postmarketing experience section of the label to include interstitial lung disease and pneumonitis.1
Sentinel analyses rely on data that can be consistently structured, traced back to clinical events, and reanalyzed as questions evolve. That same expectation now extends beyond postmarket safety surveillance.
How real-world evidence supports FDA approvals
FDA approvals that incorporate real-world evidence reflect similar discipline around data preparation and study design.
For Actemra, FDA relied in part on a randomized controlled trial that used national death records to assess 28-day mortality. Incorporating real-world mortality data strengthened the completeness of the primary endpoint, showing how external data sources can be integrated into traditional trial designs when standard follow-up alone is not sufficient.1
In rare disease, FDA has increasingly relied on well-curated natural history data to fill gaps where randomized comparators are difficult or impossible to enroll. The approval of Voxzogo for achondroplasia combined evidence from a randomized study with external controls drawn from a multicenter natural history registry. That approach allowed FDA to evaluate treatment effects against real-world disease progression, grounded in longitudinal patient data collected outside of the trial setting.1
Registry-based evidence has also informed decisions in complex care settings. For Orencia, FDA considered real-world evidence from a non-interventional study using transplant registry data to evaluate survival outcomes following hematopoietic stem cell transplantation. Here again, the value came from longitudinal data captured in routine care and structured in a way that supported regulatory review.1
A regulatory shift that lowers the barrier to using RWE
Recent FDA guidance signals that this approach to real-world evidence will be emphasized moving forward.3 In December 2025, the agency removed a long-standing requirement that identifiable, patient-level data must always be submitted when real-world evidence is used in certain medical device applications, and indicated that similar updates are under consideration for drugs and biologics.3
Historically, this expectation limited the practical use of many large, high-value data sources, even when they captured meaningful real-world outcomes at scale. Under the updated guidance, FDA reviewers will evaluate the strength of submitted real-world evidence on a case-by-case basis, recognizing that valid regulatory insight can be generated from de-identified and aggregated data when it is fit for purpose.
This change opens the door to broader use of national registries, claims databases, and electronic health record networks similar to the types of data environments that have powered Sentinel safety studies and real-world regulatory decisions to date. It also reinforces a clear signal to sponsors: the FDA is focused less on rigid submission mechanics and more on whether the evidence can reliably answer the regulatory question at hand.
As covered in Fierce Biotech, this could lead to the use of far larger, de-identified real-world datasets in regulatory reviews, expanding how evidence from registries, claims, and electronic health records informs FDA decisions.4
Real-world evidence matters now because the FDA is actively using it to inform safety and approval decisions, raising the bar for data readiness and making regulatory-grade RWE a practical requirement, not a future aspiration.
Stay ahead of new regulations and expectations: explore how HealthVerity Marketplace supports regulatory-ready real-world evidence, with curated data built for FDA safety surveillance and decision-making.
References
- Use of real-world evidence in regulatory decision making. FDA. Published online September 26, 2025. Accessed December 15, 2025. https://www.fda.gov/science-research/real-world-evidence/fda-use-real-world-evidence-regulatory-decision-making
- Healthverity being evaluated for fda’s sentinel initiative. HealthVerity. March 20, 2023. Accessed December 15, 2025. https://healthverity.com/news/healthverity-being-evaluated-for-fdas-sentinel-initiative/
- FDA eliminates major barrier to using real-world evidence in drug and device application reviews. FDA. December 15, 2025. Accessed December 15, 2025. https://www.fda.gov/news-events/press-announcements/fda-eliminates-major-barrier-using-real-world-evidence-drug-and-device-application-reviews
- Hale C. FDA opens door to more real-world datasets in device submissions. December 16, 2025. Accessed December 16, 2025. https://www.fiercebiotech.com/medtech/fda-opens-door-massive-real-world-datasets-medical-device-submissions
