Real-World Evidence in Medical Device Submissions: FDA's Current Acceptance Criteria

Real-World Evidence in Medical Device Submissions: What FDA Actually Expects

Real-world evidence (RWE) has moved from a regulatory buzzword to a legitimate submission tool -- but only when it meets FDA's increasingly specific expectations. For device companies trying to reduce clinical trial burden or support a post-market label expansion, understanding exactly where the agency draws the line between acceptable RWE and inadequate observational data is not optional. It is the difference between a substantive review cycle and a major deficiency letter.

This post breaks down FDA's current framework for RWE acceptance in medical device submissions, grounded in the guidance documents and statutory authority that govern your submission strategy today.

The Statutory and Regulatory Foundation

FDA's authority to accept real-world data (RWD) and real-world evidence in device reviews derives from Section 505F of the Federal Food, Drug, and Cosmetic Act, added by the 21st Century Cures Act in 2016. While originally written with drugs in mind, the agency has extended this framework to devices through a series of guidance documents and pilot programs.

The foundational device-specific document is the August 2017 guidance, 'Use of Real-World Evidence to Support Regulatory Decision-Making for Medical Devices.' This guidance remains the primary reference for any sponsor considering an RWE strategy. It is not a green light for using registry data instead of a clinical trial. It is a framework that outlines when and how RWD can generate credible evidence that meets the evidentiary standards under 21 CFR Part 860 for PMAs, or the substantial equivalence standard under Section 513(i)(1)(A) of the FD&C Act for 510(k)s.

What FDA Means by 'Fit-for-Purpose' Data

The single most important concept in FDA's RWE framework is fit-for-purpose. The agency will not accept RWD simply because it exists in large volume. Reviewers evaluate whether the data source, collection methodology, and resulting evidence are appropriate for the specific regulatory question being answered.

Under FDA's framework, fit-for-purpose evaluation covers three dimensions:

• Data relevance: Does the RWD capture the patient population, indication, and clinical endpoints that are material to your submission? Registry data from a broad cardiovascular population is not automatically relevant to a specific structural heart device indication.
• Data reliability: Was the data collected in a manner that supports valid statistical inference? This includes completeness, accuracy, and freedom from systematic bias. FDA expects sponsors to document data governance processes and audit trails consistent with 21 CFR Part 11 where electronic records are involved.
• Analytical validity: Is the statistical methodology pre-specified and appropriate? Ad hoc analyses conducted after data access are a significant credibility problem with agency reviewers.

RWE Across Submission Pathways

510(k) Submissions

For 510(k)s, RWE is most commonly used to support performance data when bench testing alone is insufficient or when a predicate's clinical data is limited. FDA has accepted RWD from registries and electronic health records to support substantial equivalence arguments, but reviewers look carefully at whether the comparator device and patient population in the RWD match the predicate relationship being claimed. If they do not, you are inviting a request for additional clinical data that could have been anticipated in pre-submission.

PMA Submissions and Supplements

RWE plays a larger role in PMA supplements, particularly for expanded indications. FDA's 2021 action plan for the National Evaluation System for health Technology (NEST) has accelerated the infrastructure for using curated registry data in PMA supplement reviews. However, for original PMAs, FDA continues to treat prospective, controlled clinical data as the evidentiary standard. RWE may supplement but rarely replaces traditional clinical evidence at this pathway level without explicit prior agreement from the division.

De Novo Requests

De Novo submissions present an underutilized opportunity for RWE integration. Because De Novo establishes new special controls for a novel device type, well-characterized RWD can help define the performance benchmarks that become part of the special controls framework itself -- shaping the regulatory landscape for an entire device category.

Common Deficiencies FDA Identifies in RWE Submissions

Based on FDA feedback patterns and agency communications, the most frequent problems with RWE submissions include:

• Failure to pre-specify the analytical plan before accessing the dataset
• Missing or inadequate documentation of data provenance and collection methodology
• Selecting a comparator group that introduces selection bias without adequate statistical adjustment
• Using RWD collected outside the US without demonstrating generalizability to the US intended use population
• Inadequate missing data handling that is not disclosed or addressed in the statistical analysis plan

The Pre-Submission Meeting Is Non-Negotiable

If you are considering an RWE strategy, engaging FDA through a Q-submission (formerly pre-IDE or pre-submission meeting) before finalizing your approach is not a best practice -- it is a practical necessity. The agency has been explicit that sponsors should obtain feedback on their RWE framework prior to submission. Walking into a 510(k) or PMA review with an RWE package that was never discussed with the relevant division is a significant risk exposure that experienced regulatory teams do not take.

Build Your RWE Strategy on a Defensible Foundation

Real-world evidence is a powerful regulatory tool when it is deployed correctly. The companies that benefit from it are the ones that treat RWE strategy as a technical discipline -- not a shortcut. That means pre-specifying everything, documenting data quality rigorously, engaging FDA early, and understanding that the agency's acceptance criteria are more demanding than most sponsors initially expect.

At ADB Consulting and CRO Inc., Andre Butler and the team work directly with medical device startups and established manufacturers to build RWE strategies that hold up under FDA scrutiny -- from Q-submission preparation through final submission and review. If you are evaluating whether RWE is viable for your program, the time to get that question answered is before you commit resources to a data strategy that may not meet the agency's fit-for-purpose standard.

Book a free discovery call with ADB Consulting and CRO Inc. today at adbccro.com and get a straight answer on whether RWE is the right path for your submission.