Capella IRB vs the site's IRB: when you need both, which goes first, and how to keep one file consistent

Do you always need Capella's IRB, even when the hospital has its own?

Yes. Capella's Human Research Protections policy requires everyone conducting academic research under Capella program requirements to obtain Capella IRB approval before any research-related interaction with participants or their data, and the SOPs make IRB approval (or a Not Human Subjects Research determination) a milestone that every doctoral project must pass. A hospital IRB's approval does not satisfy that milestone on its own.

Could Capella rely on the other board instead? In principle the SOPs allow it: Capella's IRB "has the authority to allow an IRB from a different institution to be the IRB of Record" where Capella cannot provide adequate oversight or wants to reduce duplicate review, subject to confirming the other institution's federal assurance and signing an IRB Authorization Agreement. The federal rules on cooperative research likewise let institutions "rely on the review of another IRB, or make similar arrangements for avoiding duplication of effort." In practice, for a doctoral project, plan on two reviews: reliance is the exception Capella decides case by case. Ask about it if the site raises it; do not build a plan on it.

When does the site's IRB have to review too?

Two different tests answer this, and both matter. The first is the federal idea of engagement. OHRP treats an institution as engaged in research when its employees or agents intervene or interact with participants for the research, obtain consent, or obtain identifiable private information for research purposes. An institution whose employees merely release identifiable records to an investigator at another institution is, under OHRP's final guidance, not engaged; neither is one that simply lets you use its rooms. A hospital whose nurses will recruit patients for you, or whose staff will pull and code charts under your protocol, is engaged, and an engaged institution needs its own IRB oversight or a reliance arrangement.

The second test is the site's own policy, and it is usually the operative one. Many health systems, districts and agencies require their IRB or research council to review every study touching their patients, staff or records regardless of engagement, and Capella's policy obliges you to "comply with additional regulations or stipulations made by the research site." Capella's SOPs warn that schools, VA facilities, military sites and "other complex organizations" often require multiple permissions, and that it is "incumbent on the researcher to determine who has the authority to issue permission." So the practical rule: ask the site, in writing, whether its IRB must review, and keep the answer. If the answer is no, the site's permission letter is your document. If yes, you have two boards and an ordering problem.

Which goes first?

The site decides, and there are three patterns. Capella's SOPs require site permission "prior to final IRB approval," and the FAQ says letters must be on letterhead, signed by the appropriate signatory, and dated within six months of IRB submission; the SOPs add that the signatory may be "an IRB Chair (if applicable) or other official within the organization." The patterns below all satisfy that rule in different orders.

Whatever the pattern, note the sentence that governs all of them: Capella's SOPs say that a study approved with conditions still may begin "absolutely no research related activities, including recruitment." A conditional letter is a bridge for the site's paperwork, not permission to start.

What counts as site permission when the site has an IRB?

They are not the same thing, and the board checks both. Site permission is an authorised official's written statement that you may do the named activities with the named people or records at that organisation; Capella says implied consent "does not constitute adequate site permission." An IRB approval is an ethics board's determination. In some organisations the IRB chair also grants site access and one letter does both jobs; in many health systems the IRB approves the protocol while a nursing director, medical director or research office grants access, and you need both documents. When it is unclear whether the right person signed, Capella's SOPs allow the board to ask for organisational charts or to consult the site directly. Save everyone that pass: establish who holds the authority, and get the letter from that desk, on letterhead, naming the activity. Our site-permission-letter article covers what the letter must say in detail.

How do you keep one file consistent across two boards?

Drift is the disease of dual review. Each board reads carefully and each asks for something; the site wants its own complaint line in the consent, Capella wants its template language, the site's IRB trims the sample, Capella's reviewer asks for a recruitment change. A change accepted by one board and not filed with the other produces two studies with one title, and both boards eventually notice. Capella's rules are specific: modifications to approved research need IRB approval before implementation; substantial modifications "require that researchers obtain updated letter(s) of permission from the research site(s)," while administrative ones need written notification to the site, and cosmetic ones need neither; methodology changes go back to the school first. The discipline that keeps the file whole:

1. Keep one master protocol and one master consent, version-numbered and dated in the footer, and submit the same version to both boards. Capella's consent template is the base; the site's required additions are inserted into it, not kept in a second form.
2. Reconcile the numbers before anything is filed anywhere: sample size, sites, instruments, recruitment channels, all identical to the SMR-approved research plan.
3. When either board requests a change, make it in the master, bump the version, and file it with the other board as a modification or, before approval, as a revised attachment with a note. Nothing is implemented until both have accepted it.
4. Log every letter: who signed it, its date, the protocol version it references. A site letter that cites version 2 when Capella holds version 3 is a return.
5. Where the site's IRB requires a research agreement, memorandum of understanding or data-use agreement, send it to Capella's IRB Office at once. Capella's policy forbids researchers, mentors and faculty from signing on the university's behalf; agreements naming Capella go to legal review, and the SOPs note some data-use terms Capella cannot accept. Agreements needing only your signature still go to the IRB with the application.
6. After approval, keep both approvals current: continuing review and amendments are owed to each board on its own terms, and a lapse at either one stops the study.

This is the work our desk does for dual-review files: one master, two submissions, every reply to either board answered point by point from the same version. How the end-to-end service runs describes it; the study itself, and both boards' decisions, remain where they belong.

What returns do we see most in dual-review files?

• The site's IRB approval attached as "site permission" when the site's own policy also required an administrator's access letter.
• A consent that carries the site's clauses but lost Capella's template elements in the merge.
• A protocol that the site approved after Scientific Merit Review stamped a different one, so the Capella application now contradicts the plan.
• A data-use agreement, signed by the candidate, that names Capella in its recitals and was never seen by Capella's IRB Office.
• Recruitment begun on a conditional letter because it had the word "approval" in it.

Each of these is a file problem, foreseeable before submission, and the reason a dual-review project should be read as one document by someone who has carried the pattern before.

What to do next

If your site has its own IRB, or you are not sure whether it does, tell us the site, the activities and the population, and a consultant will map which reviews you need, in which order, and what each file must carry. We draft the letters for signature, build both submissions from one master, and answer every reply until approval; the research stays yours and each board's decision is its own. Request the free application review, or read the Capella IRB process step by step and the questions we are asked most.