Predicate Device Selection for 510(k) Submissions: How to Find and Choose the Right One

2026-07-06 · Caduvo Team

Your predicate device decides your 510(k)’s testing burden and review risk. Learn what a predicate device is, how to search the FDA database with product codes, and how to rank candidates to avoid a Not Substantially Equivalent letter.

Your 510(k) submission lives or dies by the predicate device you pick. A weak choice forces extra testing. A wrong choice earns a Not Substantially Equivalent letter. Here’s how to get it right — without burning budget on unnecessary bench work.

What a Predicate Device Means for 510(k) Submissions

A predicate device is a legally marketed device that you compare your device to in a 510(k) submission to demonstrate substantial equivalence. That equivalence is not about being identical. It’s about having the same intended use and technological characteristics that do not raise new questions of safety or effectiveness. If your device has different technological features, you can still claim equivalence as long as any differences don’t introduce new risk types, and performance data shows the device is as safe and effective.

The FDA’s 510(k) program rests on this comparison. You must show that your device’s intended use matches the predicate’s and that any design divergence falls within an acceptable performance envelope. This is what makes predicate selection a core submission decision — not a paperwork exercise.

Searching the FDA Database for Candidate Predicates

Start with product codes. Look up your device’s classification regulation (21 CFR Parts 862–892) and the associated three-letter product code. That code groups devices with similar technology and intended use, making it your best filter for relevant predicates. Plug that code into the FDA 510(k) Premarket Notification database.

Filter by clearance date: use the past 10 years. Devices cleared under current standards reduce the technology gap you must bridge. Export the result list. Then, for each candidate, open the “510(k) Summary” or “Statement” PDF. Compare the indicated use statement sentence by sentence with your own. Check the technological characteristics description. Copy the substantial equivalence discussion — it often reveals the rationale FDA accepted.

A practical note: If you find no clear predicate with the same product code, search by device name or applicant name, then note the product codes those entries carry. You may discover that a device similar to yours actually cleared under a different regulation panel, which can open a viable predicate path.

Ranking Predicates: The 5-Factor Scorecard

Not all predicates are equal. Rank candidates on five factors to find the one that minimizes review risk.

  1. Intended use overlap. Identical wording is ideal. A minor variation that does not alter the clinical purpose (e.g., “for removal of soft tissue” vs. “for resection of soft tissue”) can still work. Large differences, like adding a therapeutic claim, will fail.
  2. Technological characteristics. Compare the predicate’s design, materials, energy source, and operating principle. If your device uses a different mechanism of action, you may need a new type of performance test.
  3. Clearance recency. A predicate cleared after 2015 usually aligns with modern consensus standards and FDA’s current thinking. Older predicates may force you to fill testing and documentation gaps that didn’t exist at the time.
  4. Availability of performance data. The 510(k) Summary often includes test methods and acceptance criteria. If those are public, you can design a benchmarking study that mirrors what FDA already accepted, lowering the burden.
  5. Single vs. split-predicate feasibility. A single predicate that covers both intended use and technology is always safer. If you must use a split — one predicate for intended use, another for a specific design feature — the burden of proving equivalence rises sharply.

Score each candidate against these factors. The predicate with the highest overlap and clearest public record becomes your primary comparator. A platform like Caduvo can surface predicate networks and product code associations quickly, but the scoring judgment remains a team decision.

Split Predicates and Other High-Risk Moves

A split-predicate strategy means you reference one device for intended use and a different device for a technological characteristic. While allowable, it’s a fragile bridge. The FDA reviewer must be convinced that combining the two does not create a new intended use and that the mixed safety profile is still equivalent. Often, it triggers requests for additional clinical data because no single device has demonstrated the exact combination you’re proposing.

Example: You want to introduce a novel coating on an implant. You pick Predicate A for the implant design and Predicate B for the coating technology. But Predicate B’s coating was never used on a load-bearing implant. FDA may view the combined device as raising new questions of safety and require a clinical study. That can add 12–18 months to your timeline.

A weak predicate — one with ambiguous labeling, missing test summaries, or an unusually narrow indication — creates a different problem. Your 510(k) submission must essentially replicate the predicate’s original testing, but with no roadmap. If the predicate was cleared under a de novo or an older standard, you may end up negotiating test plans with the reviewer from scratch.

Actionable Takeaways

Related reading