The Workbench · Craft
Emergency research trades consent for a public process
This blog has already drawn the line between an IDE's emergency-use deviation, its treatment-use pathway, and the single-patient consent exception at 21 CFR 50.23 — and named, without stopping there, the fourth provision those three get confused with: 21 CFR 50.24, the exception for planned emergency research. 50.24 doesn't belong on that same list of narrow, single-patient carve-outs. It authorizes an entire study to enroll subjects without their consent, or a legally authorized representative's, and the price for that authority is a set of collective obligations most consent-focused files have no place to record at all.
The exception only reaches a specific kind of trial
21 CFR 50.24(a) sets conditions on the trial itself before consent is even discussed. Subjects have to be in a life-threatening situation where available treatments are unproven or unsatisfactory and valid scientific evidence is needed to establish safety and effectiveness. Consent has to be genuinely infeasible: the subjects can't consent because of their own condition, the intervention has to be given before a legally authorized representative could reasonably be reached, and there's no way to identify eligible subjects in advance to seek consent from them ahead of time. And the research has to hold out the prospect of direct benefit to the subjects themselves, grounded in preclinical evidence and a risk profile reasonable against the alternative of standard therapy. A protocol that can't clear all three findings together doesn't reach 50.24 at all — it isn't a lighter-touch version of ordinary consent, available whenever obtaining consent would just be inconvenient.
The IRB has to document its own findings, not just its comfort
An IRB approving research under 50.24 isn't approving a study its members happen to trust. It has to find and document, in writing, that the regulation's conditions are actually met — the same evidentiary discipline any IRB action rests on, applied here to a finding that removes the ordinary consent requirement entirely rather than approving a protocol that already has one. That documentation is the record a study file has to preserve on its own terms, because it's the IRB's finding, not the sponsor's assertion, that the exception applies at all.
Consent doesn't disappear; it moves to the community
In place of individual consent, 50.24 requires four things: consultation with representatives of the communities the trial will be conducted in and subjects drawn from; public disclosure to those communities, before the trial starts, of its plans, risks, and expected benefits; a second public disclosure after the trial completes, detailed enough — including the demographic makeup of the research population and the results — for the community and other researchers to actually evaluate what happened; and an independent data monitoring committee with standing oversight while the trial runs. None of those four substitutes for the eight elements an ordinary consent form still owes when the exception isn't in play — they're a different kind of protection, aimed at a public that's standing in for the individual consent the trial can't obtain.
A family member can still object, and a subject can still withdraw
The exception doesn't eliminate every individual-level protection just because it eliminates the initial consent requirement. The IRB has to review and approve a specific procedure for letting a family member object to a subject's participation before enrollment goes forward, and a separate procedure for telling the subject — or a legally authorized representative or family member, once one is reachable — that participation can be stopped at any point without penalty. A protocol file that records the community-consultation plan but not these two individual-level procedures has documented half of what 50.24 actually requires before a subject can be enrolled.
FDA's own concurrence sits on top of the IRB's finding
For FDA-regulated research, the IRB's finding isn't the last word either. The sponsor submits the plan to FDA, along with the community-consultation and public-disclosure materials, for FDA's own review before the exception can actually be used, and the disclosed materials are filed publicly as part of that record. A study file that stops at the IRB's approval letter is missing the federal layer of review the exception itself depends on.
Where this meets the file
A study file built around 50.24 needs each of these carried as its own tracked item — the three-part eligibility finding, the IRB's documented conditions, the community-consultation and disclosure record, the family-objection and withdrawal procedures, and FDA's own concurrence — rather than one exception flag borrowed from an ordinary consent tracker. A clinical-study consent-exception worksheet built around that structure is previewed in the launch catalog. If your program has run a study under 50.24 and tracks it 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.