‘Least burdensome’ limits the ask, not the bar

“Least burdensome” gets invoked in regulatory strategy meetings the way a favorable precedent gets invoked in a legal argument — as a reason FDA shouldn't be allowed to ask for something. That reading has the direction backwards. The least burdensome provisions are a statutory instruction to FDA about how it requests information, not a doctrine a sponsor gets to raise to withhold information a review genuinely needs. Reading it as the second thing is how a submission strategy ends up arguing with a reviewer over a principle that was never built to help it make that argument.

The instruction lives in two places in the FD&C Act: Section 513(i)(1)(D), governing what FDA can request to support a substantial equivalence determination in a 510(k), and Section 515(c)(5), the parallel instruction for a PMA's reasonable-assurance-of-safety-and-effectiveness determination. Both provisions were added by the FDA Modernization Act of 1997 and later amended by the FDA Safety and Innovation Act of 2012, and both say the same thing in different postures: when FDA requests information from a sponsor, it has to consider the least burdensome means of getting an adequate answer, and request accordingly.

Three elements, and none of them is “less”

FDA's current guidance on the provisions — The Least Burdensome Provisions: Concept and Principles, finalized February 5, 2019, superseding a version in place since October 2002 — defines least burdensome as the minimum amount of information necessary to adequately address a relevant regulatory question, through the most efficient manner, at the right time. All three clauses carry weight. “Minimum necessary” isn't “minimum plausible” — it's scoped to what actually answers the open question. “Most efficient manner” asks whether a faster or cheaper method reaches the same evidentiary answer, not whether a weaker one does. “Right time” is a scheduling instruction as much as a substantive one: a question worth raising in a Q-Submission, before a sponsor has committed resources to the wrong study design, is not least-burdensomely raised for the first time in a deficiency letter after the fact.

The standard doesn't move; only the path to it does

FDA's guidance is explicit that the least burdensome principle doesn't alter the substantial equivalence standard, the reasonable-assurance standard, or any other statutory or regulatory criterion a device has to meet. It governs how FDA gets to a determination, not what that determination requires. A sponsor that cites “least burdensome” to argue a reviewer's request for missing safety data is somehow improper has misapplied the principle to a question it was never written to answer — if the data addresses a real, still-open question about the device, the request is doing exactly what the statute contemplates, regardless of how burdensome supplying it turns out to be.

The clearest built-in example is a standard substituting for a study

The principle isn't only an instruction reviewers apply case by case; FDA has also built it directly into pathway design. An Abbreviated 510(k) lets conformance to a recognized consensus standard stand in for the underlying performance data that standard was built to generate — the most efficient available means of answering a question FDA would otherwise have to ask a sponsor to answer from scratch. That substitution is least burdensome in its purest form: the regulatory question doesn't go away, and the evidentiary bar behind the standard doesn't drop, but the sponsor isn't asked to re-derive evidence a recognized standard has already generated on the device's behalf.

It cuts against FDA staff too, not only in the sponsor's favor

The obligation runs on FDA's side of the desk as much as it shapes what a sponsor should expect. The guidance directs review staff not to request the maximum information a question could theoretically support, but the minimum that actually resolves it — which is also why a sponsor's own least-burdensome argument works best made as a scoping question during review (or earlier, in a Q-Submission) rather than as a rationale for declining to answer one already raised. A benefit-risk case built for a De Novo request runs on the same logic in reverse: the case has to be sufficient, not exhaustive, and knowing which open questions actually need answering is what keeps it from becoming both.

Where this meets the rest of the file

A submission strategy that treats least burdensome as a scoping discipline — what does this specific regulatory question actually require, and what's the most efficient way to answer it — gets more out of the provision than one that treats it as leverage against a data request already on the table. A pre-submission worksheet built around naming the open regulatory question before proposing how to close it is previewed in the launch catalog. If your team applies the least burdensome principle differently, the shelf takes that correction directly.