Most 510(k) predicate research stops at the clearance record. Device name, product code, intended use, decision date — the fields you'd find on FDA's 510(k) database. That tells you what FDA cleared and when. It doesn't tell you how the device has performed in the real world since.

The FDA's MAUDE database — Medical Device Adverse Event Reporting System — fills that gap. It contains millions of reports submitted by manufacturers, importers, and device facilities when a device may have caused or contributed to a patient death, serious injury, or malfunction. Checking MAUDE before committing to a predicate is not standard practice in the industry. It should be.

What MAUDE can tell you about a predicate

A high MAUDE report count for a product code doesn't automatically disqualify a predicate. High-volume devices generate more reports simply because they're used more. But MAUDE data can surface patterns worth investigating before you build a submission around a device:

  • Recurring malfunction types — if the same failure mode appears across dozens of reports, FDA may have updated its thinking on the design or materials involved
  • Death or serious injury reports — a predicate associated with patient deaths warrants careful review of whether those outcomes relate to the device characteristics you're relying on
  • Recent report spikes — a sudden increase in reports after years of low activity can indicate an emerging problem FDA hasn't formally addressed yet

How to use MAUDE data in predicate selection

MAUDE data is most useful as a filter during the shortlisting phase — after you've identified 3-5 candidates that pass the basic regulatory fit test (same product code, recent clearance, matching intended use) but before you've committed to one.

For each candidate, look at the adverse event record for that product code. Pay attention to:

  • Total report volume relative to other candidates in the same product code
  • Distribution across malfunction, injury, and death categories
  • Whether reports cluster around a specific failure mode that would affect your device's design
  • The trend over the last 2-3 years — improving, stable, or worsening

This isn't about finding a "clean" predicate with zero adverse events. It's about understanding whether the post-market record introduces complexity you'd need to address in your submission — or whether it signals that FDA's view of this device type has evolved since the clearance you're relying on.

When adverse event history affects your SE argument

Adverse events don't directly affect the legal validity of a predicate — a cleared device remains a valid predicate regardless of its post-market record. But they can affect the practical strength of your submission in a few ways.

If FDA has issued guidance, letters to industry, or updated special controls in response to adverse event patterns for a device type, those developments may affect what your submission needs to demonstrate. A predicate with a significant adverse event history in the same failure mode your device addresses may actually be an opportunity — your submission can argue that your design specifically mitigates the risk the predicate's record revealed.

Either way, you want to know before you file, not after FDA asks.

The practical barrier

The reason most predicate research skips MAUDE is the friction involved. FDA's MAUDE search interface is slow, requires knowing the right search fields, and doesn't connect to the 510(k) database. Running MAUDE searches for each predicate candidate as part of a normal research workflow adds significant time.

The check is worth doing even when it's slow. Building a submission around a predicate with a significant adverse event record — and finding out during FDA review — is more expensive than the time it takes to check upfront.

MAUDE in Predicase

Predicase Pro surfaces MAUDE adverse event summaries directly on every predicate page — total reports, event types, and trend — without leaving your research workflow.

Start free trial →

Everything you need for a 510(k) submission. Nothing you don't.

14-day free trial. No setup. Cancel anytime.

Start free trial →