The Workbench · Craft
MDUFA's 90 days is a goal, not a deadline
A submission tracker that logs one FDA clearance deadline — ninety days out from filing, circled on the calendar — is enforcing a number the Food, Drug, and Cosmetic Act never wrote. Section 510(k), codified at 21 U.S.C. 360(k), requires a device to clear FDA before entering commercial distribution. It says nothing at all about how long FDA has to decide. The ninety-day figure teams actually plan against comes from somewhere else entirely: the Medical Device User Fee Amendments, a negotiated goals letter FDA and industry renegotiate on a multi-year cycle, currently MDUFA V, covering fiscal years 2023 through 2027. Treating a negotiated performance target as a statutory clock misreads what happens if the date slips, and misreads whose delay actually caused it.
Section 510(k) itself is silent on timing
The statute behind a 510(k) clearance sets out what a device needs to show — substantial equivalence to a predicate — and says nothing about how many days FDA has to decide it. That's a real contrast with the PMA pathway sitting one tier up: section 515(d) of the Act does address PMA review timing directly, even though FDA regularly extends past it with the applicant's agreement. A 510(k) has no equivalent statutory anchor at all. Whatever number a team is planning a launch timeline around for a 510(k) didn't come from the law creating the pathway.
MDUFA is a fee bargain, not a rule
The number comes from the Medical Device User Fee Amendments instead — the arrangement under which industry pays fees to fund a share of FDA's device review work, and FDA commits in exchange to a set of performance goals spelled out in a negotiated goals letter. The current iteration, MDUFA V, covers fiscal years 2023 through 2027, and its 510(k) goals run in tiers: an FDA decision within 90 FDA Days for 95% of submissions, within 180 FDA Days for 90% of submissions, and — for anything still open past 100 FDA Days — written feedback identifying the outstanding issues, delivered through a meeting or teleconference rather than left unexplained. Those are commitments FDA reports its own performance against, published and renegotiated on a cycle tied to the user-fee reauthorization, not obligations Congress wrote into the statute governing the pathway itself.
“FDA Days” and calendar days are two different clocks
The 90-day figure is also narrower than it sounds. FDA's own guidance on the review clock, FDA and Industry Actions on Premarket Notification (510(k)) Submissions: Effect on FDA Review Clock and Goals, describes the clock stopping whenever a submission is placed on hold awaiting the sponsor's own response to an Additional Information request — that hold time doesn't count against FDA's 90-day figure at all. The Total Time to Decision goal MDUFA V tracks separately, by contrast, is a shared-outcome measure that does count the sponsor's own turnaround alongside FDA's review time, and lands around 112 calendar days under the current goals letter — noticeably longer than 90, for exactly that reason. A team quoting “90 days” as though it were the full elapsed time from filing to decision is quoting the shorter of two different clocks, and attributing the gap entirely to FDA when a slow response to an AI request can be the sponsor's own doing.
What this changes about a submission tracker
A tracker built around one submission date and one decision date can't tell those two clocks apart, and ends up crediting or blaming FDA for a span that includes the sponsor's own response time. What the file needs instead is the submission date, the start and end date of every AI hold, and the decision date — enough to compute FDA's own working time separately from the calendar span, and to see which side actually held the file last whenever a review runs long.
Where this meets the file
The distinction matters earlier than the review clock itself: a submission that fails Acceptance Review never starts the MDUFA clock running in the first place, whatever goal date a launch plan assumed. A submission-timeline tracker built around FDA Days, AI holds, and the Total Time to Decision goal separately — the same distinction behind the timeline estimator previewed in the tools section of the launch catalog — is built to keep those clocks apart rather than collapsing them into one deadline. If your program tracks MDUFA goals differently, the shelf takes that correction directly.
The Regulatory Toolkit launches soon — a free shelf of source-mapped templates, checklists and browser-only tools for regulatory teams. Get one email when it opens, or contribute a template.