The guidance that skipped the draft

Most CDRH guidance follows a choreography regular enough to feel like a rule: draft posts, a comment window opens, FDA reads what arrives, and a final version eventually replaces it. Assume every guidance document takes that path and you will be right most of the time — and wrong in exactly the case where being wrong costs the most, because FDA is not required to take that path at all.

The paragraph that permits skipping the draft

FDA’s Good Guidance Practices regulation, 21 CFR 10.115, sorts guidance into two tiers of significance and, for the more significant tier — Level 1 — normally calls for a draft and a comment period before the agency issues it as final. Paragraph (g) is the exception: FDA may implement a Level 1 guidance document immediately, without prior public participation, when the agency determines that prior participation is not feasible or appropriate. A guidance issued this way is final the day it posts. It was never a draft, and none of the signals a reader would watch for on the ordinary path — a comment window, a Draft title — will have appeared ahead of time.

Final does not mean the conversation is over

The regulation does not leave the public without a lever, even after the fact. 21 CFR 10.115(g)(5) states it directly: you can comment on any guidance document at any time, and FDA will consider comments received and revise the guidance when appropriate. What that lacks, next to a draft’s comment window, is the one thing a deadline provides — a date by which a comment is guaranteed a hearing before the decision gets made. On a draft, filing before the closing date buys a seat at the table before the document exists in final form. On a guidance that was final from day one, the policy is already in force; a comment afterward asks FDA to reconsider something already operative — worth doing, but a materially different position to argue from.

The label that looks the same either way

A Federal Register record for a guidance document carries a title, and the title is how “draft” gets communicated at all — Draft Guidance for Industry versus Guidance for Industry. A guidance issued under paragraph (g) skips straight to the second form. Read the record without knowing the mechanism exists and it looks identical to any other final guidance: the ordinary conclusion of a draft you might simply have missed. It is not that. It is a document that was never a draft, in effect from the date it posted — and the title alone will not tell you which of the two you are looking at. No visible comment window is not evidence that none was needed, and it is equally not evidence that one already closed.

Whether a specific guidance’s immediate effect changes anything for a specific submission is a determination for whoever is working that file, against that guidance’s actual content — this describes a mechanism in the rule, not a reading of any document in particular. FDA Radar’s Federal Register adapter tags every guidance item as draft or final from the title alone, and keeps the effective date and the comment deadline as two separate facts on every document it ingests — collapsing them into “posted, so presumably commentable” would erase exactly the distinction this piece is about.