A PMA approval letter closes out the review, not the manufacturer’s obligations. Every year after that — for as long as the approval stands — the periodic report is the one scheduled moment where a reviewer who isn’t looking at this device day to day gets to check whether it’s still the device FDA actually approved.
What the regulation actually asks for
21 CFR § 814.84 sets the content: a description of any changes to the device or its labeling made without a supplement — the kind the regulation doesn’t require prior approval for, not the kind that does — a summary of unpublished data and reports bearing on the device’s safety or effectiveness that the applicant hasn’t already reported to FDA, plus a bibliography of anything published, and, where the approval order required one, an update on any post-approval study. None of this is raw data submitted for the first time. Almost all of it should already exist somewhere in the manufacturer’s own records, in the risk management file or the complaint trending that feeds it. The periodic report is where someone has to pull those threads into one document a reviewer who isn’t inside the company can actually follow.
Why the roll-forward habit is the risk
The easiest way to produce a periodic report is to start from last year’s and update the dates. That habit is also how drift survives past the year it started: a labeling change made for a good reason in March gets summarized correctly in this year’s report and then never mentioned again, even though it’s still true two years later and worth someone re-checking against what the device now actually does. Reading last year’s report before writing this year’s — not as a template, but as a claim to re-verify — is what keeps the document doing the job it’s actually for: catching a device that has quietly moved away from its approved file before an inspector or a competitor does.
Where this goes wrong
Treating it as a template rather than a re-verification
Copying last year’s report forward with only the dates updated, so a change from two cycles ago never gets checked against what’s true now.
Reporting a change without reconciling it against the file
Naming a device or labeling change in the narrative without confirming the design history file and the current labeling actually agree with what the report says happened.
Waiting for a supplement trigger to notice labeling drift
Assuming that if a change didn’t require a supplement, it doesn’t need tracking anywhere else either — the periodic report is often the only place it’s on record at all.
None of this changes what the regulation technically requires. It changes who benefits from reading the report closely: the same people who’d be asked to explain a labeling mismatch to an inspector are the ones positioned to catch it first, in their own document, before anyone else does.
Sources & further reading
- 21 CFR § 814.84 — Periodic Reports ecfr.gov
- Regulatory Academy — How to Read an IND Annual Report regulatoryacademy.com
- Regulatory Academy — How to Read a PMA Summary of Safety and Effectiveness Data regulatoryacademy.com
- Regulatory Academy — How to Read a Risk Management File regulatoryacademy.com
This essay is provided for general educational purposes and reflects the regulatory landscape as of its publication date. It is not legal, regulatory, or career advice.