The predicate device you choose is the single most consequential early regulatory decision a 510(k) medtech makes — and the one investors scrutinize hardest. A weak or mismatched predicate doesn't just slow your clearance; it signals to a diligence team that the regulatory strategy hasn't been thought through. The good news: predicate quality is assessable from public FDA data, so you can pick one that survives scrutiny.
1 Substantial equivalence is a claim you have to defend
Your predicate must share the same intended use and technological characteristics — or any differences must not raise new questions of safety and effectiveness. Diligence teams pull your predicate's 510(k) summary from the FDA database and check whether the equivalence story actually holds.
How to close it
Write the equivalence argument before you file. If your device differs materially, name the difference and the data that addresses it.
2 A recalled or problem predicate is a red flag
If your predicate has Class I/II recalls or a heavy MAUDE adverse-event history, you inherit that scrutiny. All of this is public.
The trap
Picking the most convenient predicate without checking its recall and adverse-event record. Investors will.
3 Predicate chains and split predicates invite questions
Relying on a long chain of predicates-of-predicates, or stitching characteristics from multiple devices, is legitimate but raises reviewer questions. Be ready to defend each link.
How to close it
Map your predicate lineage explicitly and keep it as short and direct as the equivalence argument allows.
The bottom line
Predicate selection is an engineering and regulatory judgment, not a search for the easiest match. A predicate you can defend on intended use, technology, and safety record is worth more than a convenient one — and it's the difference between a clean diligence and a stalled one.
Know what an investor will find — first
We'll run a complimentary Signal Brief on your space — a short, fully-cited read drawn entirely from public sources. No cost, no pitch.