What FDA's new human factors submission categories actually sort

On May 29, 2026, FDA finalized Content of Human Factors Information in Medical Device Marketing Submissions, closing out a document that had sat in draft since December 2022. The revised report templates for both IVD and non-IVD submissions take effect August 1, 2026, and every human factors report submitted from that date has to state which of three new Human Factors Submission Categories it falls into before the rest of the report carries any weight with a reviewer. The categories sort by risk and by what a device modification actually touches, not by how long the report runs, and a sponsor still building toward the old, uncategorized outline is finishing a document FDA's own template no longer asks for.

A companion to the 2016 guidance, not a replacement

FDA is explicit that the new guidance is a companion to, not a replacement for, its 2016 guidance Applying Human Factors and Usability Engineering to Medical Devices. The 2016 document still governs the human factors engineering process itself — the analytical work, the formative evaluations, the critical-task identification, the summative validation testing under simulated-use conditions. The 2026 guidance governs a narrower, separate question: what has to appear in the report a sponsor actually hands FDA describing that work. A team that read only the 2016 guidance and assumed it covered submission content has been reading the process manual as if it were the filing instructions.

Three categories, sorted by what a modification touches

The final guidance sorts every submission into one of three Human Factors Submission Categories, set by how much risk a device or a modification to it carries and by whether that modification actually touches a critical task or a use-related hazard. Category 1 sits at the low end — a modified device that doesn't meaningfully affect critical tasks can carry a high-level summary of the human factors evaluation instead of a full validation report, an option the pre-2026 outline had no defined path for. The categories above it scale the required content up from there, and FDA added a new decision point to its existing risk-based flowchart — Decision Point D — specifically to route a submission to the category it belongs in, rather than leaving that call to the sponsor's own read of the risk.

The report's own section order changed underneath the categories

Independent of the category question, the recommended report outline itself was reordered. The prior draft structure separated the summary of formative evaluations from the use-related risk analysis and critical-task sections with other content in between; the final guidance places the formative-evaluation summary directly ahead of both, in sequence — the work that surfaced the use problems, immediately followed by the risk analysis those problems fed, immediately followed by the critical-task determination that analysis produced. It's a narrative fix, not a scientific one: the same evidence in the old order asked a reviewer to reconstruct that chain themselves; in the new order, the document does it for them.

What still traces back to 2016, unchanged

The critical-task definition itself didn't move: a task that, if performed incorrectly or not performed at all, would or could cause serious harm to the patient or user, where harm includes compromised medical care. Formative evaluation still does the job of surfacing use problems and use errors that analysis alone won't predict; summative testing still stands as the primary evidence of use safety, run under a structured, predefined protocol against critical tasks and worst-case conditions. What the 2026 guidance changed is the obligation to present that same unchanged work in a specific order, under a specific category label — not the underlying engineering.

Where this meets the rest of the file

The critical tasks a human factors report identifies are supposed to trace back to the same use-related hazards a risk management file has to carry hazard by hazard, and the usability gap the category framework is built to close is the exact one a completeness pass against an instructions-for-use checklist can't reach on its own — a document that lists every required IFU section can still leave a critical task unvalidated, which is precisely the evidence this new category system is designed to make a reviewer ask for by name.

A human factors submission built around this structure — the category determined through Decision Point D rather than assumed, the report ordered to the 2026 outline, and critical tasks traced back to the risk file underneath them — is previewed in the launch catalog. If your program's category determination works differently, the shelf takes that correction directly.