Real-world evidence at the FDA
The FDA's use of real-world evidence grew from post-market safety monitoring into a programme that bears on approvals. Laws, guidance documents and a dedicated centre mark its stages, and the agency has based labelling and approval decisions on it.
Origins in post-market safety monitoring
The FDA's use of real-world data began with post-market safety monitoring and grew into a more comprehensive approach, with the Prescription Drug User Fee Act of 2002 (PDUFA III) and the 21st Century Cures Act of 2016 as its legislative milestones. PDUFA III introduced user fees to speed drug reviews. Because electronic healthcare data were becoming a valuable source for medical research, it also included provisions for considering observational data in regulatory decisions on drug safety. The act laid the groundwork for the agency's gradual integration of real-world evidence into its regulatory framework. The Sentinel Initiative, launched in 2008, is a prime example of that integration: it uses a standard data model to combine sources of insurance claims and electronic health records to evaluate the safety of medical products after they have been approved and are in use in the general population.
The 21st Century Cures Act and effectiveness
The 21st Century Cures Act of 2016 required the FDA to establish a programme to evaluate the potential use of real-world evidence to support the approval of new indications for approved drugs or to satisfy post-approval study requirements. It also explicitly enabled the agency to consider real-world data in assessing drug effectiveness, beyond the traditional role of real-world evidence in safety monitoring. The agency's approach recognises the potential of real-world evidence to inform regulatory decisions throughout a product's lifecycle, from the approval of new indications to post-approval studies. The agency's framework for real-world evidence sets out to ensure transparency in its regulatory use, define appropriate use cases, establish standards for data quality and reliability, and develop methods for analysis and interpretation. The framework stresses data quality, study design and analytical method, and the role of stakeholder engagement in developing standards.
Guidance documents
The FDA issued several guidance documents to implement the framework, and these five are among the key ones:
- Data standards for drug and biological product submissions containing real-world data — standardised formats and requirements for real-world data in applications, to keep the evaluation of drug and biological products consistent and reliable.
- Use of real-world evidence to support regulatory decision-making for medical devices — approval, labelling changes and post-marketing surveillance of devices.
- Submitting documents using real-world data and real-world evidence to FDA for drugs and biologics — recommendations on submitting documents that contain them, focused on the evaluation of drugs and biological products using real-world information.
- Considerations for the design and conduct of externally controlled trials for drug and biological products — designing, conducting and analysing such trials for regulatory decisions.
- Real-world data: assessing electronic health records and medical claims data to support regulatory decision-making — evaluating the relevance and reliability of those two sources.
Together they set expectations on data collection and quality, including provenance, completeness and accuracy, and on study design and analysis, where the observational nature of much real-world data demands attention to bias and confounding. Their aim is real-world evidence that is reliable, relevant and appropriately integrated into the regulatory process, and the agency keeps a full list of them on its website. They also reflect the agency's commitment to an evolving healthcare system in which insights from routine clinical practice can inform regulatory decisions and the reverse, an approach that aims at a more efficient and responsive regulatory process able to keep pace with advances in medical science and technology.
To guide and encourage the use of big data in industry, the FDA issued final guidance on using real-world evidence to support regulatory decisions for medical devices in August 2017 and on using electronic health record data in clinical investigations in July 2018, and draft guidance on submitting documents that use real-world data and evidence for drugs and biologics in May 2019.
In July 2024 the FDA finalised its guidance on assessing electronic health records and medical claims data to support regulatory decision-making for drug and biological products, which clarifies study design elements, including the selection of study variables and validation.
The Center for Real-World Evidence Innovation
The Center for Real-World Evidence Innovation, established by the agency's Center for Drug Evaluation and Research, aims to promote and facilitate the use of real-world evidence in drug development and regulatory processes, making drug evaluation and approval more efficient and effective. It was created because of that evidence's growing role in drug development and the need for a coordinated approach to evaluating it and integrating it into regulatory processes. As of 2025, it was expected to significantly influence how real-world evidence is used in drug approvals, potentially making regulatory decisions more efficient and patient-centric, and to inform future policy and guidance, giving regulators and pharmaceutical companies clearer direction on designing and analysing fit-for-purpose studies.
Labelling and approval decisions
The FDA has increasingly recognised the value of real-world evidence in supporting faster, better-informed regulatory decisions, and such evidence can support drug approvals, label expansions and post-market safety monitoring. It has used real-world evidence in rare diseases and paediatric populations, where large randomised trials may be infeasible or unethical. For example, it has considered evidence from natural-history studies and patient registries to support the approval of treatments for rare genetic disorders.
Methotrexate, used in some lymphomas, breast cancer and autoimmune diseases such as rheumatoid arthritis and refractory psoriasis, is highly toxic, with severe side effects including death, especially when taken too frequently. In 2019 the FDA used Sentinel to investigate dosing errors among patients with rheumatoid arthritis newly using the drug, and estimated that 0.4% had an overdose needing rescue therapy. With adverse event reports, this finding led the agency to require changes to the prescribing information: a new warning about dosing errors, a clarified dosing schedule for non-oncological uses and patient medication guides.
Fluoroquinolones are antibiotics sometimes used for acute bacterial sinusitis, bronchitis and urinary tract infections, and because of safety concerns their label states that they should be prescribed only when patients have no alternative. Real-world studies contributed significantly to the evidence behind the boxed warnings on tendon rupture in 2008, on irreversible peripheral neuropathy in 2013 and on worsening myasthenia gravis. At a 2015 advisory committee meeting, real-world data showed the drugs widely used despite restrictive labelling. The meeting weighed real-world studies on peripheral neuropathy, retinal detachment, tendon rupture, cardiac arrhythmia and aortic aneurysm, which complemented trial data and spontaneous reports to give a fuller picture of the drugs' safety in real-world use and led to multiple regulatory actions to evaluate their risks and benefits.
In 2019 Ibrance (palbociclib) received a supplemental approval for male breast cancer based on trial information supplemented by real-world evidence derived from electronic health records in the Flatiron Health database, which showed safety and effectiveness in men consistent with those in women. A 2025 account credits the real-world study with expanding use, and so patients' access, faster than trials could have, and reads the decision as regulatory flexibility and as showing that large record databases can yield regulatory-grade evidence, especially for rare conditions and under-represented groups.
In July 2021 Prograf (tacrolimus) was approved for preventing organ rejection in lung transplantation based solely on real-world evidence. The evidence came from the US Scientific Registry of Transplant Recipients, comparing tacrolimus-based with cyclosporine-based immunosuppression in adults and children on patient and graft survival at one year. Outcomes, including one-year survival, were better with Prograf, and its safety was consistent with that seen in other transplant populations. The same 2025 account reads the decision as expanding treatment options for lung transplant recipients, setting a precedent for high-quality registry data in approvals and validating the potential of real-world data to inform regulatory decisions.