Predicate Devices: How to Find and Choose One for Your 510(k)

2026-08-11 · Caduvo Team

How to find and choose a predicate device for your 510(k) submission, including a practical FDA database search workflow, ranking criteria, and risks of weak or split predicates.

Your 510(k) submission lives or dies by the predicate device you choose. Pick the wrong one and you’ll add months of testing, increase your risk of a Refuse to Accept letter, or force a last‑minute pivot to De Novo. Pick the right one and the clearance path becomes a checklist of familiar evidence.

A predicate device is not just a reference. It’s the benchmark against which the FDA judges your entire submission. This post walks through how to locate strong candidates, how to rank them, and how to avoid the predicate traps that routinely stall 510(k) reviews.

What Is a Predicate Device?

A predicate device is a legally marketed device to which you claim substantial equivalence in a 510(k). To receive 510(k) clearance, your device must have the same intended use and, if there are technological differences, you must show those differences do not raise new questions of safety or effectiveness.

The predicate sets the bar. Everything that follows in your submission—biocompatibility, mechanical testing, software validation—is measured against what the predicate required or what the FDA believes is needed to match it.

You are not required to use a single predicate. You can reference multiple predicates (a “split predicate”) to cover different features, but that strategy adds complexity. We’ll address the risks later.

How to Search for Predicate Devices in FDA Databases

The FDA’s 510(k) Premarket Notification database is the starting point. Use a focused workflow to pull candidates that match your device’s intended use and technology.

1. Identify your device’s likely regulation number. Look up similar devices in the FDA classification database (medical device classification) to find the appropriate 21 CFR part (e.g., 21 CFR 870.1250 for a vascular clamp).
2. Search the FDA Product Classification database using that regulation number or keywords. Note the associated three‑letter product codes (e.g., DWS for an external pacemaker).
3. Open the 510(k) Premarket Notification database. Search by product code. Filter by decision date—last 10 years is a practical window. You’ll get a list of cleared devices with links to 510(k) Summaries.
4. Read the Summaries. Look for indications for use, technological description, and the specific testing performed. A detailed Summary often means there is enough public information to replicate the testing package.

If you can’t find a predicate with a matching product code, broaden your keyword search. But a mismatch in product code often signals a different classification or intended use and will invite additional FDA questions.

Ranking Predicate Candidates: What Makes a Good Predicate?

Not all predicates are equal. Rank your candidates on the criteria below. A high‑quality predicate reduces testing burden and lowers Review By Exception risk.

• Same intended use / indications for use. The closer the intended patient population, anatomy, and disease state, the fewer bridging arguments you must write.
• Technological overlap. Same fundamental operating principle, materials, design, and energy source? If not, list every difference and estimate what additional performance data you’ll need to show equivalent safety.
• Testing burden. Review the predicate’s 510(k) Summary for the testing it referenced. If it relied on recognized consensus standards (e.g., ISO 10993 for biocompatibility, IEC 60601 for safety), you can often follow the same path. If it required a non‑clinical animal study or clinical data, your burden scales accordingly.
• Recency. A predicate cleared in the last 5–7 years reflects current review expectations. Older devices may be grandfathered on standards the FDA no longer accepts.
• Clean safety history. Search the MAUDE database. Avoid predicates with recent Class I recalls or a pattern of adverse events linked to the device’s design.
• Product code alignment. Using a predicate from the exact same product code means the FDA will classify your device identically. That eliminates one classification argument upfront.

A strong predicate is one where you can write a straightforward comparison table and point to testing your team can reliably execute. If you are already sketching out a dozen differences, keep searching.

The Risks of a Weak or Split Predicate Strategy

A weak predicate—one with significantly different indications, a higher device class, or an outdated technology—forces you to generate extra data to bridge the gaps. The FDA can issue an NSE (Not Substantially Equivalent) determination, which sends you to the De Novo pathway or back to the drawing board.

A split predicate approach, where you use two or more predicates to piece together the intended use and technology, is allowable but risky. The FDA will scrutinize whether the combination truly represents a single device. They may ask you to identify a primary predicate and justify why the secondary predicate’s features do not introduce new risks. Split predicates often lengthen the review and can lead to multiple rounds of Additional Information requests.

If your device genuinely combines features that have never been cleared together and no single predicate exists, a split predicate may be your only route. In that case, start the conversation with the FDA early—via a Pre‑Submission—to align on the plan before you commit to testing.

Actionable Takeaways

• Start with product codes, not device names. A three‑letter code narrows your search to devices classified the same way yours will be.
• Focus on recent 510(k) Summaries that spell out detailed testing protocols. If a Summary is thin, the predicate may not give you enough of a roadmap.
• Score each candidate on three dimensions: intended use, technology, and testing burden. The candidate with the fewest differences almost always saves you time.
• Avoid split predicates unless no single device fits. If you must split, lock in the primary predicate early and use the Pre‑Submission program to pressure‑test the strategy.

An hour spent ranking predicates with rigor can save months of avoidable testing and rewrite cycles during review. Treat your predicate search as the first design control decision—not an afterthought.