The recognition number isn't the list number

A Declaration of Conformity form asks a submitter to list, for every consensus standard it invokes, the standard’s own recognition number — the identifier FDA assigned when it recognized that standard. Scan the Federal Register notices that make those recognitions happen and a different number shows up in almost every title: Recognition List Number: 066, the most recent as of this writing. Both are FDA numbers, both sit next to the word “recognition,” and they identify two unrelated things.

One standard, one number

FDA keeps every standard it currently recognizes — in whole, in part, or not at all — in a single searchable database, the FDA Recognized Consensus Standards Database. A standard that clears recognition gets its own number there, tied to that standard, that edition, and whatever scope FDA recognized it for. Appropriate Use of Voluntary Consensus Standards in Premarket Submissions for Medical Devices, FDA’s current guidance on using the database in a submission, is the document that tells a submitter to copy that number onto the Declaration of Conformity form, next to the standard it belongs to.

One batch, a different number

FDA doesn’t recognize standards one at a time in public. Additions, withdrawals, corrections, and revisions accumulate and go out together, a few times a year, as a Federal Register notice titled “Modifications to the List of Recognized Standards” — each one carrying its own sequential Recognition List Number. List 065 published February 19, 2026; List 066 followed August 24, 2026. The list number identifies that specific notice and everything it changed. It says nothing about any single standard inside it — two completely unrelated standards updated in the same notice share the same list number and nothing else.

A recognition number names a standard. A Recognition List number names a Tuesday’s worth of paperwork. Copying the second one onto a Declaration of Conformity answers a question nobody asked.

What actually takes effect, and when

FDA’s own recognition-list notices have used the same language for years: the modifications described are incorporated into the database, and the recognition becomes effective upon publication in the Federal Register — not whenever a reader happens to notice the database changed. Recognition and Withdrawal of Voluntary Consensus Standards, FDA’s guidance governing how recognition and withdrawal actually work, sits behind that sequence. The database and the Federal Register notice move together, as two faces of a single act, which is also why a standard’s withdrawal is dated to a specific list number — not to whenever a user of the database happened to look.

The gap a numbered list doesn’t cover

The numbered lists are not the only way this database changes. Beginning with Recognition List 033, published January 14, 2014, FDA stopped separately announcing minor revisions through the numbered Federal Register process at all — those changes now land only in the database itself, outside any list number a reader could watch for. A team whose standards-monitoring consists of checking each numbered Recognition List notice as it publishes has a real feed for everything that goes through that channel, and no feed at all for a minor revision that doesn’t.

The practice

Before a Declaration of Conformity goes into a submission, the number that belongs on it is the one attached to the standard itself, at the edition and scope the database currently shows — confirmed against a live search, not against whichever Recognition List notice a team happened to read last. A standard FDA has withdrawn, never recognized, or recognized with a narrower scope than the one being claimed doesn’t qualify for the declaration at all; the fallback is the full-data “general use” route, not a form with the wrong number on it.

FDA Radar doesn’t monitor the Recognized Consensus Standards database today — see what we monitor and what we deliberately don’t, which says so directly. Until a reliable feed exists, reading the live database before a submission goes out is a manual check, not an automated one, and which standards and editions a specific submission needs is a determination for the regulatory professional on that file, never regulatory advice from this column.