How long does GCU IRB approval take? Count gates and passes, not a clock

It is the first question almost every candidate asks, and the only honest answer is that it is the wrong unit. A Grand Canyon University file does not sit in a queue accumulating elapsed anything. It clears a set of gates, and then it makes a number of passes across a desk. Both of those are countable. Neither converts into a date anyone can promise you.

Caroline Strauss, PhD · 2026-08-23

The short answer

No one can honestly quote a date, because approval is not a waiting period. It is a count of gates cleared before the board reads and passes the file makes afterwards. Almost every gate sits under your control; almost none of the passes do.

Why will nobody give you a number?

Because three of the four variables belong to somebody else, and the fourth is unknowable until the file is read.

The first variable is the level of review, assigned rather than chosen: GCU processes protocols as exempt, expedited or full board, and a file read by a designated reviewer travels a different road from one going to the convened board. The second is the board's own calendar, which belongs to the board. GCU's IRB materials have described convened-board submissions as tied to a published cut-off rather than a rolling intake — and that cut-off, like the forms, is set by GCU's current doctoral handbook or IRB portal.

The third variable is the number of returns, and this is the one that dominates every file. A protocol that comes back with modifications required has not lost a place in a queue; it has acquired an extra pass, and the pass after that one depends on what your response does. The fourth is the state of the file when it is filed — how many of the gates below were actually clear at the moment of submission, as opposed to nearly clear.

A practice that quotes you a date is quoting a guess about someone else's desk. We do not. What we can do is count the gates with you, and say honestly which of them are open.

What are the gates before the board reads anything?

At GCU the IRB is the last desk, not the first. The proposal reaches it through the committee and chair inside the DC Network and then through academic quality review, and the board reads the application against the proposal that review approved. Each gate below has to be clear before the application is genuinely readable.

The committee and chair. The manuscript has to be settled at the level of method before anything downstream can be built against it. Nothing filed against an unsettled proposal survives.
Academic quality review. The approved proposal is the text the board reads back against. Filing an application without it gives the board nothing to compare — a point covered in full in the process in order.
Training on every name. GCU requires CITI human-subjects training and the responsible conduct of research course, registered with the GCU address so that completions load into the portal automatically. A lapsed or mismatched certificate is a stall in the plainest sense: nothing else in the file can be read past it.
Chair approval on the submission itself. GCU's own preparation steps place the chair's approval before the application goes in, not after.
Site authorization in hand. The board requires an authorization or cooperation letter from each site, and GCU states that the letter travels with the submission, before any activity is conducted at that site. A site you intend to approach later is a gate that is currently shut.
Instruments with their provenance. The instrument exactly as approved, with permission or license attached. A renamed or lightly revised instrument re-opens a gate two desks back.
The application and every attachment, assembled. GCU's submission runs through its current web-based protocol system, with the attachments loaded rather than promised.

Read that list again and notice what it is. Not a waiting list: every item is an action, and every one is yours or your chair's. That is precisely why honest advice about the pace of an IRB file is almost entirely about the part before filing.

What are the passes once the file is in?

A pass is one crossing of the board's desk. The count of passes, not any elapsed measure, is the thing a good file actually reduces.

The completeness read. Before anything substantive, the file is checked for the pieces it must contain. A file missing a letter or a certificate does not enter review; it is handed back.
The level determination. The board places the protocol against the federal criteria as exempt, expedited or full board. This is a decision issued to you, never a status you assert — the reasoning is set out in the levels of review.
The substantive read. A designated reviewer for minimal-risk work; the convened board where a protected population or greater than minimal risk is involved. The convened route adds a cut-off to the count, because a convened body reads on its own schedule.
The letter. Approved, approved with modifications required, deferred pending further information, or — rarely — not approvable as submitted. The board's own labels sit in GCU's current IRB portal.
The response pass. Every point answered in the board's order, with each changed sentence changed identically everywhere it appears in the file.
The re-read. One more pass, and then either the letter or another list of points.

That is the whole machine. Passes one through four happen once. Five and six repeat until the board is satisfied, and the count of repetitions is where files diverge most dramatically. Two candidates who filed on the same afternoon can finish in entirely different places, and the difference is almost never luck.

Which gate stalls a GCU file most often?

Drift — daylight between the application and the proposal academic quality review approved. It is not an ethics failure and it is rarely deliberate. A sample is resized after a committee conversation. An instrument is renamed. A consent form promises confidentiality that the procedures cannot keep because a linking sheet exists. Each of those is a sentence in one document contradicting a sentence in another, and each one buys an extra pass.

Behind drift, in rough order of frequency: a site named in the procedures with no letter attached; a letter that authorizes less than the application describes; a certificate that lapsed between proposal and submission; an instrument with no permission; and a data-security plan that describes identifiers three different ways in three different places. Each of these is a gate you can close before filing, and none of them requires anybody else's calendar.

There is a sixth, and it is the cruellest: a response to the board that quietly rewrites an approved procedure in order to satisfy a point. That is how a file earns its second return, and how a second return earns a third.

What can you actually control?

Each gate, who holds it, and what a stall looks like
Gate	Whose hand is on it	What opens it	What a stall looks like
Settled method	You and your committee	Chair and committee satisfied with the manuscript	A method still being argued while documents are drafted
Approved proposal	Academic quality review	The approval itself	An application written against a draft nobody has approved
Training current	You	Completions recorded against every name on the protocol	A certificate that lapsed while the proposal moved
Site authorization	The site	A signed letter matching the procedures exactly	A site that will "confirm later"; a signature without authority
Instruments and permissions	You and the instrument's owner	The approved version with license attached	A revised instrument nobody re-approved
Level of review	The board	The board's determination	A file arguing for a level rather than earning it
Convened reading	The board	The board's own cut-off and agenda	Nothing you can act on; plan around it, do not chase it
The response	You	Every point answered, no new drift introduced	A patch that changes one document and not the others

A map of the gates, not a schedule. Nothing here predicts or promises a board decision, and no consultant can shorten a board's own reading.

Does external review change the count?

Yes, and it is the change candidates most often fail to plan for. Where a hospital, a district or an employer runs its own review, a second body joins the sequence with its own gates, its own submission requirements and its own agenda — and the two boards frequently disagree about which of them should read first. That is a distinct problem with a distinct solution, and it is worth reading before you approach the site rather than after.

So what does a consultant actually change?

Not the board's calendar, and not the board's decision. Those are the board's and stay the board's. What can be changed is the count: how many gates are genuinely clear at the moment of filing, and how many passes the file makes before the letter. A file built against the approved proposal, with every letter gathered and every document saying the same thing as every other, is a file with fewer things to come back about. That is the entire mechanism, and it is why the first move on this desk is a reading rather than a draft.

What to do next

Take the gate list above and mark each one open or shut, honestly. Any gate you cannot mark open is the real answer to your question about pace — and nearly always one you can close yourself, without waiting on anybody's permission. Send us the AQR-approved proposal and whatever the board has already said, and the free application review will tell you which gates are shut and what closes them. More reasoning sits in the answers on the record.

Request the free application review or take up the letter that stalls most files

Sources

Where this article and GCU's own materials differ, GCU's current doctoral handbook or IRB portal governs. gcuirb is an independent practice and is not affiliated with, endorsed by or connected to Grand Canyon University.