The 2026 transition
What changes for ISO 11607 on 20 December 2026
A premarket submission in your pipeline is the only thing this date can touch. Find out whether yours is affected, what you would change if it is, and how to verify every dated claim here against FDA's own record in minutes.
In short
Which recognition covers which edition, and until when
FDA recognition 14-530 covers ISO 11607-1:2019 without Amendment 1, and declarations of conformity to it are accepted in premarket submissions until 20 December 2026. Recognition 14-594 covers the amended edition. Amendment 1:2023 applies risk management to packaging as a defined process, so this is a documentation gap rather than a testing gap. It reaches premarket submissions only, not devices already on the market.
Before you check anything
The two ideas that let you read your own declaration, and check every claim here
FDA recognizes particular editions of particular consensus standards. A recognition number identifies one of those entries: a standard, an edition, and the extent of recognition given to it. The number is not a property of the standard. It is a property of FDA's record about that edition of it.
A declaration of conformity in a premarket submission is a statement that the device meets a recognized standard, offered so that conformity to the standard can carry that part of the submission rather than the underlying data alone. That is why the edition is the thing that matters here. The declaration points at a recognition entry, and the entry names the edition.
Conformity is voluntary. A manufacturer can meet the requirements without declaring conformity to any standard and provide the data instead. What is not optional is that a declaration you do make is accurate about what it points at.
Two consequences follow, and both are true of ISO 11607-1 right now. A recognition can be superseded while the standard on your shelf does not move. And a superseded recognition can remain usable for declarations until a date the database states, rather than stopping the day its successor appears.
FIG 01
The transition, and the one thing in your file it turns on
Every row here turns on one thing: which edition your declaration of conformity can point at in a premarket submission. None of it touches testing, and none of it touches a device already on the market.
| Recognition 14-530 | ISO 11607-1:2019, without Amendment 1 |
| Declarations to 14-530 | Accepted in premarket submissions until 20 December 2026 |
| Recognition 14-594 | Supersedes 14-530, for the edition including Amendment 1:2023 |
| What actually changes | Which edition a declaration of conformity can point at |
| Who it reaches | Premarket submissions filed after 20 December 2026 |
| Who it does not reach | Devices already legally marketed, with no submission pending |
| Tests added by the transition | None |
| Data invalidated by the transition | None |
| ISO 11607-2 | A separate part, with its own recognition entries: 14-531 and 14-595, read on 4 August 2026. Declarations to 14-531 are accepted in premarket submissions until the same date, 20 December 2026 |
The four entries behind the dates on this page, 14-530 and 14-594 for Part 1 and 14-531 and 14-595 for Part 2, were read in the FDA Recognized Consensus Standards database on 4 August 2026 and a copy of each was kept. The database is linked further down so you can read them yourself. If our page and that database disagree, the database is right and we want to hear about it.
The nature of the gap
A documentation gap, not a testing gap
The transition changes which edition a declaration can point at. It does not add a test, move an acceptance criterion, or expire data you already hold. A sterile barrier system that met the requirements in 2023 has not become worse since.
Amendment 1:2023 applies risk management to packaging as a defined process. So the work that closes the gap is writing rather than running: showing where the packaging decisions came from, and how they connect to the risk management file the manufacturer already maintains under ISO 14971.
That is a cheaper class of work than testing, and it is far cheaper before a submission than inside an Additional Information response, where the same writing has to be done under a clock with a reviewer waiting. It is also work an in-house team can do. Nothing about this transition requires an outside firm.
The one thing the date cannot tell you about your file
Whether writing is all your file needs is a question about your file, not about the date. If a packaging decision was never made on a documented basis, then writing it up afterwards is not the same thing as having made it that way, and pretending otherwise is how a documentation exercise turns into a finding. The date does not tell you which of those two situations you are in. Reading the file does.
The substance
What Amendment 1:2023 asks you to show in your packaging file
The amendment's subject is the application of risk management to the packaging system. Not risk management in general, and not a paragraph asserting that risk was considered somewhere: a process, applied to packaging decisions, with the outputs recorded where a reader can follow them.
What that looks like in a file is a visible connection running both ways between the packaging documentation and the ISO 14971 risk management file. What could go wrong with the sterile barrier system, what the packaging design and the validation do about it, and where the evidence for that sits. A reader should be able to start at a risk and arrive at a test result, or start at a test result and arrive at the risk it answers.
Our reading, labeled as ours: a packaging file assembled before 2024 had no particular reason to show that linkage, because the edition it was written against did not ask for it. Nothing in such a file was wrong when it was written. The text moved, and the file did not move with it. That is a different situation from having done the work badly, and it should be described that way to whoever signs off the remediation.
Why this page describes the amendment and does not quote it
We cite a clause only from the text of the standard, read here. ISO 11607-1 and ISO 11607-2 are paid documents and we have not bought either, so no clause number from either one appears on this page. That includes the amendment, which is described at the level our sources support. What this page does cite is the regulator's own record: four recognition entries, read in FDA's database on 4 August 2026, each one linked and checkable in under a minute. A page about document control that cited a clause nobody here has read would be a strange thing to publish.
In your own file
Seven things to look for before you decide this affects you
None of these needs a consultant and none of them needs a quotation. They are in the order that gets you to an answer fastest, which means the first one may end the exercise on its own.
Find the declaration, or the draft of it. Does it name ISO 11607-1, an edition year, and a recognition number? A declaration that names the standard but not the edition is ambiguous about the one thing this transition is about, and it is ambiguous today, before any deadline.
Checked in our file
Protocols, reports and the packaging validation plan carry a standards section, and that section records the edition the work was done against. A mismatch between what the declaration claims and what the evidence cites shows up there first.
Checked in our file
Take a single packaging decision, a material, a seal parameter, a shelf life claim, and try to trace it to an entry in the risk management file and back again. If that round trip cannot be made on paper, that is the gap the amendment is about, in the most concrete form it takes.
Checked in our file
Does the risk management file mention the sterile barrier system at all? Two files can each assume the other holds the linkage, and between them hold nothing. That is worth ten minutes with both documents open before anyone concludes the packaging file is the problem.
Checked in our file
Check every standard the packaging file cites against the edition FDA recognizes now, not against the edition that was current when the file was written. It is a public database, the check is free, and it covers every standard the file cites rather than 11607 alone.
Checked in our file
The work is documentation, so it belongs to whoever owns the packaging file and whoever owns the risk file, together. Where that is two people who have never had a reason to sit down over it, the constraint is calendar time rather than budget, and it is worth booking early for that reason alone.
Checked in our file
Check us
How to read the recognition entries for yourself
Every dated statement on this page comes out of one public database. It is free, it needs no account, and the whole check takes a few minutes. We would rather you did it than took our word for it. Once you have read an entry for yourself, an undated claim about a recognized edition, from us or from anybody else, is easy to see through.
The FDA Recognized Consensus Standards database, linked below. It is the agency's own published record of which editions it recognizes, which is why it settles a question that a summary of it cannot.
Checked in our file
Search on the standard rather than on the recognition number. The result set is small enough to read in full, and reading it in full is what shows you the supersession relationships instead of one entry in isolation. It also puts the Part 1 and the Part 2 entries in front of you at the same time.
Checked in our file
Each entry names the recognition number, the standard and edition it covers, the extent of recognition, and the dates attached to it. The date of interest here is the one governing how long declarations to that recognition are accepted.
Checked in our file
One covers ISO 11607-1:2019 without Amendment 1 and carries the date this page is about. The other covers the edition including Amendment 1:2023. Seeing them side by side is the fastest way to understand that this is a change of edition rather than a change of requirement.
Checked in our file
ISO 11607-2 has its own entries in that same result set: 14-531, and 14-595 for the edition including the amendment. We read them on 4 August 2026 and the Part 2 date is the same as the Part 1 date, 20 December 2026. Read the entry rather than assuming that, which is the whole point of the step. The entry states the date; the parallel does not.
Checked in our file
If the database and this page disagree, the database is right. The page is corrected on this URL, the correction is dated and stays visible rather than being edited away, and you get a personal reply from us, not a form letter.
Checked in our file
The FDA Recognized Consensus Standards database
Free, public, and the record every dated statement on this page came out of. It is the same link the methodology page carries, on purpose: one source, cited the same way wherever it appears on this site.
The answers and the notes stay in this browser. Nothing is sent anywhere, there is no account, and nothing about this checklist is collected. Print the page if you want to keep them.
The other half
What it does not mean, stated as plainly as what it does
- It is not a recall and it is not an enforcement date
Nothing is being withdrawn, and no product becomes non-compliant on 20 December 2026 by operation of this transition. What ends is the availability of one recognition for a declaration of conformity in a new submission.
- It does not reach devices already on the market
A legally marketed device with no submission pending is not affected by this date. Anyone describing a December cliff for cleared products is scaring their reader inaccurately, and that is worth remembering about the source as well as about the claim.
- It does not require anyone to revalidate anything
The transition adds no test and invalidates no data. Whether something in a particular file needs redoing is a question about that file, and the date does not answer it in either direction.
- It does not make conformity mandatory
Conformity to a recognized consensus standard is voluntary. The transition changes what a declaration can point at. It does not change whether you have to make one, and providing the data directly remains available.
- It is not a statement about what ISO 11607-2 requires
Everything on this page is about Part 1, apart from the Part 2 dates, which come from the Part 2 recognition entries and say nothing about content. Sealing process validation sits in the text of Part 2, we have not bought that standard, and nothing here should be read as telling you what it asks for.
- It is not a reason to buy anything
The seven checks above are free and can be run in an afternoon by someone who already knows the file. If your file shows the linkage, you are finished, and nothing further follows from the date.
- It is not urgent for its own sake
The date binds submissions, so whatever pressure it creates comes from a submission timeline rather than from the calendar. A firm with nothing planned until 2028 has no deadline attached to whatever documentation work its file turns out to need, and it may turn out to need none.
What we have not verified
Three things this page deliberately does not tell you
A page written to be forwarded should be as clear about its holes as about its content, because the reader at the other end of the forward cannot ask us anything. These are ours, as of the review date at the top of the page.
- What either part of ISO 11607 actually says
ISO 11607-1 and ISO 11607-2 are paid documents and we have not bought them. Everything dated on this page comes from FDA's recognition entries, which settle which edition is recognized and until when, and settle nothing about what the text requires. So you will find no clause number from either part anywhere on this site, including the clause a lot of articles quote for the separation of stability and performance testing. What closes this one is the purchased standard, read here, with the wording transcribed from it and the date recorded.
- What happens to a submission already under review on 20 December 2026
We have not read FDA's own account of how a submission already in the queue is handled. So we do not answer it here, and we do not answer it on a call either. What would close it is a statement from the agency, read at source and recorded with the date it was read.
- A clause number for the amendment's risk management requirements
The description of Amendment 1:2023 on this page is supported by our sources. The clause-level citation is not in our standards register yet, and we do not publish a clause we have not read, however easy it would be to copy one from somebody else's article.
If you would rather someone else read it
Having the file read is a separate decision from the date
Run the seven checks above with your own team, find the linkage is there, and you are finished: nothing more follows from the date, and nothing changes hands. Those are the same checks a reading starts from. The difference is that a reading is us going through your actual file, where this page is published information rather than a look at yours.
Where a fixed-fee reading earns its keep is narrower than the deadline makes it sound: nobody in the building has time before the submission date, or the person who wrote the original file has left and the file has to speak for itself now, or a protocol is about to be run and it is cheaper to have it read before the samples ship than after the results are in.
The Packaging Standards Gap Assessment is $2,500 fixed, ten business days from receipt of documents, and involves no laboratory testing. A possible finding is that you have no gap and need nothing, and the fee is the same either way.
The Packaging Standards Gap AssessmentQuestion by question
The questions this date raises, starting with the number itself
- What is FDA recognition number 14-530?
It is FDA's recognition entry for ISO 11607-1:2019 without Amendment 1. Declarations of conformity to that recognition are accepted in premarket submissions until 20 December 2026.
- What is FDA recognition number 14-594?
It is the recognition that supersedes 14-530, covering the edition of ISO 11607-1 including Amendment 1:2023. Both entries are readable in the FDA Recognized Consensus Standards database, which is linked above.
- When does FDA stop accepting ISO 11607-1:2019 without the amendment?
Declarations of conformity to recognition 14-530 are accepted in premarket submissions until 20 December 2026. After that, the recognition covering the amended edition is the one a declaration points at. We have not read FDA's account of how a submission already under review on that date is handled, so we do not answer that part.
- Do we need to revalidate our packaging for Amendment 1:2023?
Not because of the transition. It adds no test and invalidates no data. It is a documentation matter: showing that packaging decisions were made through a risk management process and recording where that shows. If a file turns out to need more than writing, that is a fact about the file rather than about the date.
- Does the December 2026 transition affect devices already on the market?
No. It concerns what a premarket submission declares conformity to. A legally marketed device with no submission pending is not affected, and anyone saying otherwise is worth a second look.
- What does application of risk management mean in ISO 11607?
A defined process rather than an assertion. Packaging decisions are made, and shown to have been made, through risk management, with the outputs traceable to the ISO 14971 risk management file in both directions. We describe it at that level because the clause-level citation is not in our standards register yet.
- Does Amendment 1 also apply to ISO 11607-2?
ISO 11607-2 has its own FDA recognition entries and they answer half of this. 14-531 covers the 2019 edition, 14-595 covers the edition including Amendment 1:2023 and supersedes it, and declarations of conformity to 14-531 are accepted in premarket submissions until 20 December 2026. We read both on 4 August 2026. What those entries do not tell you, and what we therefore do not state, is what the amendment asks for in Part 2: that is in the text of the standard, and we have not bought it. Search 11607 in the database and all four entries sit in the same result set.
- How should ISO 11607 be cited in a declaration of conformity after December 2026?
A declaration points at a recognition entry, so our reading, labeled as ours, is that it should name the standard, the edition including the amendment, and the recognition number for that entry, taken from the database at the time of writing rather than copied from an older submission. We have not read an FDA instruction setting out the wording, so that is a position rather than a requirement. What any regulator accepts in any submission is that regulator's decision, and nobody can promise it.
- What happens if a standard is updated after we submit our declaration of conformity?
We have not read FDA's own account of that, so we do not answer it. The nearest entry on the open questions list on our methodology page is what happens to a premarket submission already under review on 20 December 2026, and it is open for the same reason. Either leaves that list by being answered from a primary source rather than by being guessed at plausibly.
- Is ISO 11607 mandatory?
No. Conformity to a recognized consensus standard is voluntary, and a declaration of conformity is one route to showing a device meets the requirements rather than the only one. That is true before and after 20 December 2026.
- We have no submission planned. Does any of this apply to us?
No. The transition governs declarations of conformity in premarket submissions. With no submission planned and no packaging change, this date does not reach you, and that is the answer you will get on a call as well as on this page.
Next
Thirty minutes, your submission date, no slides
Bring the device description, the sterilization method and the date you expect to file. The first thing you will hear is whether this transition touches you at all. If it does not, that is where the call ends, and being told so is a legitimate result of asking.
Or email directly. A personal reply the same business day, from us, not a queue.
Book a 30-minute call