Once an Institutional Review Board (IRB) has approved a research protocol, nothing about that study can change without the IRB knowing about it and, in most cases, approving the change before it takes effect. That prospective-approval requirement is the core rule under 45 CFR 46.108: except when necessary to eliminate an apparent immediate hazard to participants, an IRB must review and approve any proposed change to previously approved human subjects research — including a revised consent document — before that change is implemented. This guide walks through what counts as an amendment, how the submission and review process actually works, how long it takes, and the mistakes that most often trip up study teams.
What counts as a protocol amendment
A protocol amendment is a planned, prospective change to a protocol that has already received IRB approval — a new procedure, an added or removed study arm, a revised eligibility criterion, a change to the consent form, a new investigator, additional study sites, or a change in how data is collected or stored. The defining feature is timing: an amendment is proposed, reviewed, and approved before it happens. That is what separates an amendment from a protocol deviation, which is discovered and documented only after an unplanned departure from the approved protocol has already occurred, and from an administrative change, which is a minor logistical edit — ICH E6(R2)’s own examples are a change of monitor or a change of telephone number — that does not affect participant safety, participant rights, or the scientific integrity of the study and so does not require the same prospective full review.
Common triggers for an amendment include: adding a study procedure or assessment; changing inclusion/exclusion criteria; adding or changing a data collection instrument; revising compensation or the consent form’s risk/benefit language; adding a co-investigator or study site; changing the sample size or statistical plan; or extending recruitment to a new population. If you are unsure whether a specific change is substantive enough to require a full amendment versus a lighter-touch administrative update, your IRB office or HRPP (Human Research Protections Program) is the authority that makes that call — not the study team unilaterally, and not this guide.
Step-by-step: how to submit an amendment
1. Draft the amendment request and revise the protocol
Prepare a clean, clearly-marked description of exactly what is changing and why. Most IRBs require a tracked-changes (redlined) version of the protocol alongside a clean version, plus a short amendment summary or cover memo describing the rationale, the specific sections affected, and whether the change affects risk level, participant burden, or the consent process. If the change affects what participants are told, a revised consent form (also typically submitted in both redlined and clean form) is required alongside the protocol changes.
2. Submit through your IRB’s amendment pathway
Nearly all IRBs now manage amendments through an electronic protocol-management system (Cayuse, IRBNet, Click, iRIS, and similar platforms are common) rather than paper submission. The amendment is typically submitted against the existing approved protocol record, not as a new application, and the system usually prompts for the same core information: what is changing, why, and whether the change is being requested because of new information (e.g., a safety signal) or simply a planned protocol refinement.
3. IRB review: expedited vs. full board
How the amendment gets reviewed depends on what it changes, not on how the original study was reviewed. Under 45 CFR 46.110 / 21 CFR 56.110, minor changes to previously approved research are one of the enumerated categories eligible for expedited review, which can be carried out by the IRB chair personally or by one or more experienced reviewers the chair designates — without waiting for a convened meeting. An expedited reviewer (including the chair) can approve, require modifications, or take no action, but cannot disapprove an amendment; disapproval requires full-board review at a convened meeting per 45 CFR 46.108(b). If the amendment introduces new risk, materially changes the risk/benefit balance, or the underlying study is not itself eligible for expedited handling, the amendment goes to the full convened board instead.
4. Respond to IRB questions or requested revisions
It is routine for an IRB to come back with clarifying questions or required edits before approving an amendment, particularly for consent-language changes. Build time for at least one round of revision into your planning.
5. Receive approval and implement — not before
The change cannot be put into practice until the IRB’s written approval is in hand. This includes using a revised consent form with new participants, or applying a new eligibility criterion to screening. Implementing an unapproved change, even a well-intentioned one, is itself a reportable event at most institutions.
How long does an IRB amendment take?
There is no fixed regulatory turnaround time for either review pathway — timelines are set by institutional practice, not by the Common Rule or FDA regulations. As a general pattern reported consistently across institutional IRB/HRPP offices: expedited review of an amendment commonly takes on the order of days to roughly one to two weeks, since a single designated reviewer can take it up without waiting for a scheduled meeting. Full-board review is slower, typically weeks to a month or more, because it is tied to the IRB’s periodic meeting calendar (commonly monthly) and a pre-meeting submission deadline that the amendment has to be received by. Build the applicable timeline into study planning, especially for time-sensitive changes like adding a site or responding to a funder deadline.
The immediate-hazard exception
The one circumstance where a change can be implemented before IRB approval is when it is necessary to eliminate an apparent immediate hazard to participants. 45 CFR 46.108 carves this out explicitly, and FDA’s parallel drug-trial regulation, 21 CFR 312.30(b)(2)(ii), works the same way: a protocol change intended to eliminate an apparent immediate hazard may be implemented right away, provided FDA is subsequently notified by protocol amendment and the reviewing IRB is notified per 21 CFR 56.104(c). This is a narrow safety-valve provision, not a general license to implement changes first and seek approval later — it applies specifically to eliminating an immediate hazard, and the subsequent notification obligation still applies.
Does an amendment ever require re-consent of already-enrolled participants?
Sometimes. Under 45 CFR 46.116(b)(5), a consent process must disclose, where appropriate, that participants will be informed of significant new findings developed during the research that may relate to their willingness to continue — the regulatory basis for going back to already-enrolled participants when a material amendment changes the risk/benefit picture. The IRB, not the study team, determines whether a given amendment triggers full re-consent of currently enrolled participants, a simpler notification, or no participant-facing action at all. Build this question into every amendment submission that touches risk, benefit, or study procedures participants are already undergoing.
Amendments to FDA-regulated (drug and device) research
For research conducted under an Investigational New Drug (IND) application, ICH E6(R2) Section 1.60 defines a protocol amendment as “a written description of a change(s) to, or formal clarification of, a protocol,” and Section 4.5.2 states the investigator’s obligation directly: no deviation from, or change to, the protocol should be implemented without documented IRB/IEC approval of an amendment, except where necessary to eliminate an immediate hazard. In practice this runs in parallel to the Common Rule process described above: the same prospective-approval principle, the same immediate-hazard carve-out, with the added layer that FDA itself (not only the IRB) may need to be notified depending on what changed. If your study is FDA-regulated, confirm with your IRB and regulatory affairs office whether a given amendment also requires an IND amendment submission to FDA, separate from IRB approval.
Common pitfalls
- Implementing before approval. Even routine-seeming changes (a new screening question, an added survey) need approval first, absent the immediate-hazard exception.
- Submitting an unmarked protocol. Most IRBs need a redlined version to review efficiently; a clean-only submission slows review and invites requests for clarification.
- Forgetting the consent form. If the change affects what participants experience or are told, the consent document usually needs to change too — and needs its own redline.
- Assuming expedited eligibility. Whether an amendment qualifies for expedited review is an IRB determination based on the nature of the change and the underlying study, not something the study team can assume or self-certify.
- Missing the re-consent question. Failing to flag whether currently enrolled participants need to be informed of, or re-consented to, a material change.
- Confusing an amendment with continuing review. An amendment is a change to what the study does; continuing review is the IRB’s periodic reassessment of ongoing approval, required at intervals appropriate to risk under 45 CFR 46.109(e). The two are separate obligations and can happen independently of each other, though study teams sometimes bundle a planned amendment with an upcoming continuing review submission for efficiency.
Amendment vs. administrative change vs. deviation vs. continuing review
These four terms get confused often enough to be worth stating plainly side by side:
- Amendment: a planned, prospective change requiring IRB review and approval before implementation.
- Administrative change: a minor, purely logistical edit (e.g., a personnel phone number) that does not affect safety, rights, or scientific integrity, and does not require the same prospective full review.
- Protocol deviation: an unplanned departure from the approved protocol, discovered and documented after the fact, not proposed and approved in advance.
- Continuing review: the IRB’s periodic reassessment of a study’s ongoing approval status, on a schedule set by risk level, independent of whether any amendment has been submitted.
