Usage Examples
- The sNDA used an EHR-derived external control arm as the RWE supporting the new indication.
- FDA wants the fit-for-purpose assessment of the claims dataset before it will look at our RWE.
- That registry analysis is RWE, so it goes in the clinical data section, not the literature summary.
What is Real-World Evidence (RWE)?
Real-World Evidence is clinical evidence about a medical product's usage, benefits, or risks derived from data generated outside traditional clinical trials, such as health records, claims, and registries, rather than from a protocol-driven randomized study.
Real-World Evidence exists because a randomized trial answers a narrow question in a narrow population, and then the product meets everyone else. Congress wrote that gap into statute: section 505F of the FD&C Act directs FDA to evaluate using evidence drawn from sources other than traditional clinical trials to support new indications for approved drugs and to satisfy postapproval study requirements.
Real-World Evidence covers conclusions drawn from routinely collected health data: electronic health records, claims and billing records, product and disease registries, pharmacy dispensing, and wearable or app data. Out of scope is the data itself, which is real-world data, and any design label. RWE is an evidence source, not a lower evidentiary standard, and not a substitute for a prespecified analysis.
Real-World Evidence is applied by fitting it into an existing regulatory container. An external control arm built from EHR data must still satisfy the adequate and well-controlled criteria in 21 CFR 314.126, where historical controls are reserved for special circumstances. Registry analyses land in the clinical data section of the application. Postapproval commitments report through the annual report.
Not to be confused with
- Real-world data (RWD)
- RWD is the raw input: the EHR extract, the claims file, the registry export. RWE is the conclusion a defensible analysis of that input supports. A data licence buys RWD and produces no RWE on its own.
- External control arm
- an external control is one design that can be built from real-world data. RWE is the broader evidence category; the external control is a specific comparison structure that must still clear 21 CFR 314.126(b)(2)(v).
- Pharmacovigilance
- pharmacovigilance is a continuous, timeline-bound safety reporting obligation that runs whether or not you have a question. RWE is a purpose-built analysis answering one prespecified question, which may or may not concern safety.
- Substantial evidence
- substantial evidence is the legal standard an application must meet. RWE is a possible input to that standard, never a replacement for it; the adequate and well-controlled criteria apply either way.
There is no standalone RWE regulation. The obligations come from the statute that authorizes RWE and the application rules any RWE-based claim has to satisfy.
What you must do
- 1Confine statutory RWE reliance to the two purposes Congress authorized: supporting approval of a new indication for a drug already approved, and supporting or satisfying postapproval study requirements21 U.S.C. 355g
- 2Design any RWE study offered as evidence of effectiveness so it distinguishes the drug effect from spontaneous change in the disease, placebo effect, or biased observation21 CFR 314.126
- 3Justify an external or historical control on the record, because historical control designs are usually reserved for special circumstances such as diseases with high and predictable mortality21 CFR 314.126(b)(2)(v)
- 4Include in the clinical data section a description and analysis of any other data or information relevant to safety and effectiveness obtained from any source, foreign or domestic, which sweeps in real-world analyses you commissioned21 CFR 314.50(d)(5)(iv)
- 5Report the status of each postmarketing study commitment annually until FDA states in writing that it is fulfilled, no longer feasible, or would no longer provide useful information21 CFR 314.81(b)(2)(vii)
Common mistakes
Treating a data licence as evidence
Buying access to a claims or EHR database produces real-world data, not real-world evidence. Without a prespecified protocol, an analysis plan, and a fit-for-purpose assessment of the source, the output is a summary no reviewer can replicate, and six figures of data spend buys nothing that advances the application.
Assuming RWE lowers the evidentiary bar
An external control built from real-world data is still judged against 21 CFR 314.126, where historical controls are usually reserved for special circumstances. Sponsors who commit to an RWE-based efficacy claim without pre-agreeing the design with FDA typically find out at a review meeting, after the data is bought and the analysis is done.
Analysing first and specifying afterwards
Real-world datasets are large enough that repeated cuts will eventually produce a favourable result. A comparison specified after seeing the data cannot distinguish drug effect from biased observation, which is the exact failure 21 CFR 314.126(a) names, and reviewers discount it accordingly.
When This Matters
- The sNDA used an EHR-derived external control arm as the RWE supporting the new indication.
- FDA wants the fit-for-purpose assessment of the claims dataset before it will look at our RWE.
- That registry analysis is RWE, so it goes in the clinical data section, not the literature summary.
Frequently Asked Questions
Rarely, and only where a randomized trial is not feasible or ethical. FDA's regulations reserve historical control designs for special circumstances, such as diseases with high and predictable mortality or where the drug effect is self-evident. For most efficacy claims, RWE supports or supplements a randomized trial rather than substituting for it.
Related Use Cases
- Pharma Use Cases
Cut NDA and sNDA prep time by 60% with AI-assisted drafting and automated readiness checks
- Biotech Use Cases
Compress IND prep from 8-12 weeks to under 3 weeks with AI-assisted drafting and validation
- Clinical Workflows
Track clinical trial regulations across global markets
- Regulatory Affairs Workflows
Cut regulatory intelligence tracking from 10+ hours/week to automated, real-time alerts
Related Regulatory Intelligence
Related Actions
- Lifecycle Management
Track every sequence operation, amendment, and supplement across the product lifecycle.
- Pharma Regulatory Workflows
NDA, BLA, IND, ANDA, and supplement execution in one workspace.
- Regulatory Intelligence
Monitor FDA, EMA, ICH, and Health Canada guidance changes across active programs.
