The FDA RWE Gap: Why Your Evidence Strategy Is Already Outdated
What Medical Affairs Leaders Need to Know About FDA’s Shifting Real-World Evidence Posture – & Three Moves to Make Before the Next Guidance Lands
On December 15, 2025, FDA announced it would accept real-world evidence (RWE) from de-identified databases without requiring patient-level identifiable data in certain marketing applications. The finalized guidance covers medical devices. But the bigger signal is what comes next: FDA explicitly stated its intent to extend this policy to drugs and biologics.
For nearly a decade, the single largest practical barrier to RWE adoption wasn’t scientific rigor or regulatory skepticism – it was a data access requirement. FDA expected identifiable, individual patient-level data in submissions. Most large healthcare databases – national registries, hospital system aggregates, insurance claims datasets – operate on de-identified data by design. The requirement effectively disqualified the richest sources of real-world evidence from regulatory use.
That barrier is being removed. And the 35-drug statistic from the last nine years is about to look like a relic.
What the Regulatory Record Already Tells Us
Medical Affairs teams often treat RWE as something FDA is “warming up to.” The approval packages tell a different story. References to real-world evidence already appear across 197 unique drugs in FDA and EMA regulatory documents. External control arms – real-world comparator data used instead of a traditional control group – feature across 136 drugs, nearly evenly split between FDA and EMA. Natural history studies, where real-world disease progression data provides context for evaluating treatment effect, appear across 674 drugs.
These aren’t pilot programs. They’re established regulatory practice. The question isn’t whether FDA accepts RWE – it’s whether your evidence generation plan reflects the standard they’re already applying.
This is where competitive intelligence comes in. The approval packages of competitor and comparator products contain FDA reviewers’ own documented assessments of real-world evidence: what external control data they found persuasive, what natural history studies they referenced when evaluating efficacy, and what methodological concerns they flagged. If a competitor submitted an external control arm for a similar indication, and FDA found it insufficient, you need to know why before you design your own study. That intelligence is sitting in publicly accessible review documents.
The Post-Market Side Matters More Than You Think
There’s a second dimension to FDA’s RWE infrastructure that affects med affairs directly, and most teams don’t see it until it’s too late.
FDA’s post-market safety apparatus already runs on real-world data. FAERS, the adverse event reporting system, is embedded in the regulatory documentation of over 1,200 drugs. But FAERS is just the signal detection layer. Behind it, FDA operates a formal evaluation process for each new approval: the ARIA Sufficiency Assessment. FDA’s epidemiologists evaluate whether the Sentinel system – the agency’s active surveillance network built on claims data covering approximately 139 million lives – can adequately monitor a drug’s post-market safety. When Sentinel is sufficient, surveillance happens at population scale with no additional burden on the sponsor. When it falls short, FDA mandates a post-marketing requirement: typically, a patient registry or formal observational study, funded and operated by the sponsor for years.
The common reasons Sentinel falls short are predictable: rare diseases with too few patients in claims databases, safety outcomes that require lab values claims data doesn’t capture, follow-up periods that exceed the two-to-three-year enrollment window typical of Sentinel data partners, and pregnancy outcomes where mother-infant record linkage remains incomplete across the distributed database.
For medical affairs, the outcome of this assessment shapes your post-launch reality: budget, timelines, publication strategy, and the scope of med affairs involvement in post-approval evidence generation. And when FAERS signals escalate far enough, they trigger Drug Safety Communications – formal public alerts that appear across 125 drugs in the regulatory record. A single DSC can change your label language, reshape your medical communications, and reframe how KOLs talk about your product overnight. The SGLT2 inhibitor class saw FAERS cases of Fournier’s gangrene led to a class-wide labeling change. The GLP-1 class saw a related dynamic with gallbladder disease, with warnings added incrementally across products over time. Med affairs teams monitoring the same data streams FDA monitors are better positioned to anticipate these inflection points than teams that react after the fact.
Three Things to Do Now
1. Build your RWE competitive intelligence.
Review the approval packages of comparable products in your therapeutic area. Identify what RWE FDA evaluated, what they accepted, and what they pushed back on. This is the most efficient way to calibrate your own evidence generation design to regulatory expectations – and it’s intelligence most competitors aren’t gathering systematically.
2. Anticipate your post-approval surveillance profile.
If your product has significant safety questions, FDA will assess whether Sentinel can handle monitoring. Rare disease, long-term safety endpoints, pregnancy, and lab-dependent outcomes are consistent drivers of Sentinel insufficiency – and therefore of mandatory PMRs. Understanding this before approval gives you a head start on registry planning and resource allocation.
3. Start building de-identified data partnerships.
National registries, aggregated hospital datasets, and large claims databases are about to become more accessible for regulatory use. De-identified data was already appearing in regulatory documents across 82 drugs before the December 2025 announcement — the regulatory conversation predates the policy. When the updated drug and biologic guidance lands, you want data access agreements and study designs ready, not a cold start.
The RWE landscape has shifted. The regulatory record confirms it. The teams that adapt their evidence strategy now will have a meaningful advantage over those that wait for the guidance to tell them what they should have already been doing.
Regulatory document analysis conducted using the Basil Systems platform’s Regulatory / Approval Search module – full text search across FDA and EMA approval packages, review documents, and correspondence.