The Clutch Brief: the approval issue

Get IRB approval on the first try: the checklist boards wish you used.

2026-07-13 · BY LENA MARSH — PhD, former university IRB member

Boards rarely reject science. They return paperwork. After years of reading submissions from the other side of the table, I can tell you most delays trace to a handful of completely preventable mismatches. This checklist removes them before your protocol ever reaches a reviewer.

Understand what the board is actually reading for.

An IRB reviewer is not grading your research ambition. They are verifying that risks to participants are identified and minimized, that consent is genuinely informed, and that your documents agree with each other. That last one surprises people most: a protocol that says interviews will take thirty minutes while the consent form says forty-five reads, to a reviewer, as a study whose author is not in control of it. Precision, in this genre, is the entire performance.

Reviewers also read in a fixed order of suspicion: vulnerable populations, identifiable data, deception, recruitment pressure. If your design touches any of those, address it head-on in the protocol with your safeguards, because a risk you name and mitigate is routine while a risk the reviewer discovers is a revision letter.

Know your review category before you write.

Exempt, expedited and full board review are different roads with different paperwork and different timelines, and the category is determined by your risk profile and data practices, not by how important the study is. Anonymous surveys of adults on non-sensitive topics live at one end; interventions and protected populations live at the other.

Design toward the lightest category your question honestly permits. Small choices move you across the line: collecting names you do not need, recording video when audio suffices, recruiting through a channel that could feel coercive. Strip what the question does not require and you often strip a whole tier of review with it.

Submission timing is a strategy too. Boards work in cycles, and a complete packet arriving just ahead of a review date clears faster than perfect science dropped into a queue that closed yesterday. Ask the coordinator for the meeting calendar, and ask early, because coordinators are the most underused people in this entire process.

The submission checklist itself.

Print this, and do not submit until every line survives:

If revisions come anyway, answer like a professional.

A revision request is not a rejection, it is a negotiation opening, and the format of your response decides how fast it closes. Reply with a point-by-point letter: quote each reviewer comment, state your change, and cite exactly where in which document the change lives. Never make silent edits beyond what was requested, because unexplained differences between versions restart the reading.

Most first-round returns are consent-form wording, missing site letters and internal mismatches, which is to say they are preventable with the checklist above. If you would rather have the packet built to pass before it goes in, our IRB and research support team assembles protocols on your board's own template and drafts the revision letters when boards push back, quote free in two hours.

Questions students actually ask.

How long does IRB approval take?

It varies by board, category and season, so ask your IRB office for current turnaround expectations rather than trusting forum anecdotes. What you control is the number of review rounds. A clean, internally consistent packet that anticipates risk questions frequently clears in one cycle, while each revision round adds a full queue-and-review loop to your timeline.

What is the most common reason IRB submissions get returned?

Internal inconsistency. The protocol, consent form and recruitment materials describe the same study three slightly different ways, and reviewers cannot approve a moving target. Second place goes to missing attachments, especially site permission letters and instrument permissions. Both causes are administrative rather than scientific, which is why a final cross-document check catches most of them.

Do I need CITI training before submitting?

Almost universally yes, for every listed member of the research team, and boards check that certificates are current before review begins. Complete the human subjects modules early, because an expired or missing certificate can stall an otherwise perfect submission at intake. Keep the completion reports as files, since most systems ask you to upload them with the packet.

Can quality-improvement projects skip the IRB?

Not on your own judgment. Many QI projects are not human subjects research and receive a non-research determination, but that call belongs to the board or its determination process, not the project author. Submit the determination form, keep the letter, and your project is protected; skip the step and publication or program sign-off can unravel over a missing piece of paper.

Is it worth contacting the IRB before I submit?

Almost always. Coordinators routinely answer template, category and consent-wording questions before review, and a ten-minute conversation prevents the classic wrong-form resubmission that costs a full cycle. Bring specific questions rather than the whole design, and write down the answers, because guidance you can reference in a cover letter smooths the review that follows.

Keep going.

IRB approval help →IRB and research support →Dissertation help →
Lena Marsh
Written by
Lena Marsh
PhD, former university IRB member · On the GradClutch bench, meet the whole roster on the team page.
This is what the Clutch Brief reads like.

One email every Sunday. Deadline math, LMS quirks, rubric speedruns. Free forever.

Get the brief →
Want the packet built to pass round one? Send the draft. 2h.Get a free quote →
Online now