For most US medical devices, the regulatory path is 510(k) — a pre-market submission claiming the device is "substantially equivalent" to an already-cleared predicate device. About 95% of cleared devices go this route (vs the ~5% that require de novo or PMA pathways).
The first half of a 510(k) is finding the right predicate. This sounds easy. It is not.
Product code is everything
The FDA classifies devices into ~6,000 product codes (3-letter codes like "DXY" for hand-held electronic stethoscope). Each code represents a specific intended use + technology combination, and predicates within the same product code are far easier to defend as "substantially equivalent".
Picking the right product code is the single most important early decision in 510(k) strategy. The code determines: which predicates you can cite, which class (I, II, or III) the device falls under, what special controls apply, and even whether 510(k) is the right pathway at all.
The FDA Product Code Classification database has the official list. Reading 6,000 codes to find yours is the kind of task tools should help with.
Date matters more than people realize
A predicate cleared in 2010 is technically still valid as a predicate today. In practice, citing a 14-year-old predicate is weaker than citing a 2-year-old one. The FDA's expectations have evolved (cybersecurity, software requirements, biocompatibility) and an old predicate may not have satisfied current expectations.
Best practice: cite a predicate cleared within the last 5 years, ideally in the last 2 years. Legacy predicates that are extremely stable (basic surgical instruments, simple bandages) are an exception.
The substantially-equivalent test
Substantial equivalence has three prongs:
- Same intended use as the predicate
- Same technological characteristics, OR different characteristics that don't raise new questions of safety/effectiveness
- No new safety/effectiveness questions raised by the differences
A perfect predicate matches all three trivially. A weaker predicate forces you to argue prong 2 or 3 in detail, with comparative testing data. The right predicate is one where SE is obvious.
Sometimes you cite multiple predicates: a "primary predicate" for intended use and a "reference predicate" for a specific technology component. This is allowed and often necessary.
Side-by-side comparison is what the FDA actually wants
The body of a 510(k) submission includes a substantial equivalence comparison table — your device side-by-side with the predicate(s) on intended use, indications, technology, performance, safety, biocompatibility. The FDA reviewer reads this table to decide.
Most online 510(k) databases show one device at a time. Building the SE comparison table requires opening multiple browser tabs, copying fields manually, and hoping you didn't miss anything. This is a productivity tool that should exist.
Our FDA 510(k) predicate finder runs live searches against openFDA (the FDA's public API), filters by device name, applicant, product code, K-number, or decision date, and provides side-by-side comparison of up to 4 devices at once with field-level diff highlighting. Direct links to the official FDA record and 510(k) summary PDF for every result.