Does a Chamberlain DNP project need IRB approval? QI by design, and the determination you still file for
For most Chamberlain DNP projects the honest answer is: probably not a full IRB review, and Chamberlain says so itself. Its handbook states that most DNP projects will meet the prescreening criteria for a determination that the project is not human subjects research, and so most will never submit the full application package. But "probably not" is not a document, and the boards do not accept it. You file to obtain the determination in writing — and you should know the three published triggers that would move your project across the line, because a site board reading the same pages may apply them differently.
Usually not a full IRB application. Chamberlain's handbook expects most DNP projects to clear prescreening as work that is not human subjects research. You still file the Prescreening Review Form in NR-705 — the determination is the document that proves it.
What makes something "research" in the first place?
The federal definition is short and does a great deal of work. Research means a systematic investigation — development, testing and evaluation included — aimed at producing or adding to generalizable knowledge. The human subject definition reaches a living person whenever an investigator gathers information about them by intervening or interacting and then uses or analyses what was gathered, or else obtains, uses or generates identifiable private information about them.
Two independent conditions, then. An activity is regulated human subjects research only when it is research and it involves human subjects. Much improvement work fails the first test. Some de-identified analysis fails the second. Chamberlain's IRB Handbook quotes both definitions directly and then states the operative conclusion: projects that do not involve human subjects and do not develop or contribute to generalizable knowledge do not require IRB review.
How does Chamberlain define "generalizable knowledge"?
This is the part worth reading twice, because the federal regulations decline to define it and Chamberlain does. Chamberlain's IRB Handbook defines generalizable knowledge as information applicable to populations outside the study population — used to draw conclusions, expand theory or a field's knowledge base, or inform policy beyond the study setting.
That definition is doing something specific. It locates the question in the scope of the claim you intend to make, not in the sophistication of your methods. Run a rigorous statistical analysis on your own unit in order to change your own unit, and the claim stays local. Run the identical analysis in order to say something about units in general, and it does not.
The handbook also disposes of the anxiety behind most of the questions we receive. Some assume that disseminating improvement findings beyond the project setting turns them into a contribution to generalizable knowledge. Chamberlain notes the federal clarification that dissemination does not, by itself, make improvement work research — and federal guidance adds the converse, that intent to publish is an insufficient criterion either way. Design is what the board reads.
What are Chamberlain's three research triggers?
Chamberlain's IRB Handbook names three circumstances in which it considers a project to be research. They are unusually concrete, and they are the checklist to run against your own design before anything is drafted.
- The aim is to test a theoretical model or to assess its applicability to a particular setting.
- The project implements an intervention that is untested, or that deviates substantively from the evidence base.
- The aim is to replicate or extend a previous research study.
Trigger two is the one that catches DNP projects, for an honourable reason. A practice-change project is supposed to apply evidence that already exists. The moment your intervention is a novel combination, or is carried into a population the evidence base does not cover, you are generating knowledge rather than applying it. That is a good project. It is also, on Chamberlain's criteria, research.
Can the same project be QI to one board and research to another?
Yes, and this is the central fact of a two-board project. Federal guidance is explicit that improvement activities are sometimes designed to accomplish a research purpose as well as improve care — the two are not mutually exclusive categories, and an activity can genuinely be both. Each board applies the definition using its own people, its own population and its own liability.
So Chamberlain's prescreening determination binds Chamberlain. It is not a passport. A health system reading the same pages beside its own patients may reach the other conclusion and convene a review, and neither board is wrong. This is why we write the consent materials and the data-handling plan to the stricter of the two readings, and why the two files must tell one story — and it is the argument for opening the site's board before the Chamberlain file closes.
What do Chamberlain's own worked examples show?
The IRB Handbook publishes six worked examples with its determination on each. The pattern they teach is sharper than any abstract rule.
| The project | Chamberlain's determination | Why |
|---|---|---|
| Follow-up telephone calls to wound-care clinic clients to catch infections earlier, a programme already used effectively in other ambulatory clinics | Not research; no IRB review required | Application of existing evidence in a local setting |
| Group appointments introduced for Hmong patients in a diabetes clinic, where the literature supports group appointments for Hispanic and White patients | Research; IRB review required | The intervention is untested in this population |
| An instructor examines whether self-efficacy is influenced by level of acculturation in a particular group | Research; IRB review required | Theory testing and expansion |
| An instructor introduces a flipped-classroom format in one course section and monitors outcomes | Not research; no IRB review required | Application of current evidence in a limited setting |
| The same flipped-classroom change, plus a survey of the class whose results will appear in a research poster | Research; IRB review required | Data are collected and analysed by interacting with living people |
| Components of two well-tested remediation models combined into one comprehensive model and piloted | Research; IRB review required | Substantive deviation from the evidence base |
Paraphrased from the worked examples in Chamberlain's IRB Handbook. The determinations are Chamberlain's; your project's determination is made on your own file.
Compare rows four and five: the practice change is identical, and what moves it across the line is a survey administered to living people and analysed for a claim aimed outward. Rows one and two teach the same lesson in clinical dress — one intervention is application in one population and investigation in another, because the evidence base reaches one and not the other.
Which parts of a DNP project most often move it across the line?
Four, in our experience, and each is usually added late and casually.
- A survey or interview bolted on to an operational change. The change itself may be plainly local. The instrument interacts with living people to produce data, and it invites the question of what the data are for.
- A comparison group. The moment one unit gets the intervention so that it can be compared against another, the design has begun to look like a test of the intervention rather than an implementation of it.
- A population the evidence base does not cover. Chamberlain's own second example is exactly this, and it is the most common honest surprise.
- Identifiable patient data. This one may not change Chamberlain's determination at all, and very frequently changes the site's. It is the line most site boards will not read past.
None of these makes a project worse; they may make it better. They simply need recognising before drafting, because they change what must be written and which review path applies — taken up in the article on how review paths get assigned.
If it isn't research, why file anything at all?
Because "not research" is a finding, and findings have owners. You are not entitled to make this one about your own project — Chamberlain's IRB Handbook directs anyone unclear on whether a project constitutes research to contact the IRB rather than decide privately.
There are practical reasons too. Your faculty member must approve each iteration of the prescreening form, so the determination is coursework, not paperwork. Chamberlain's handbook bars recruitment until the written outcome is received and faculty notified. Your site will often ask what the school concluded. And Chamberlain's IRB Handbook is explicit that the board will not grant retroactive approval — so the determination must exist before the project starts, not once a manuscript needs one.
A further gate is worth knowing: Chamberlain's decision tree notes that all proposed research projects must first submit the IER Research Project Review Form, and the IRB Handbook records that the board will not act as a relying IRB where Chamberlain people or Chamberlain data are the subject. If your project studies Chamberlain itself, the route is different from the one described here.
What to do next
Test your project against the three triggers today. Is any part testing a model? Is the intervention untested in the population you will deliver it to? Are you replicating or extending a study? If all three answers are no, you likely have a practice-change project — and you still file for the determination that says so. If any answer is yes, or only partly, better to know now, while the design can absorb it.
Then request the free application review. We read the project against both tracks and send back a written assessment: how each board is likely to read it, what the prescreening file needs, and whether your site's board will accept the same framing. More questions of this kind are answered on the FAQ. Each board reaches its own determination and the project remains yours; our part is making sure the file says what the project really is.
Sources
- Chamberlain's board handbook: the definitions it adopts, its reading of generalizable knowledge, its three research criteria, the six worked examples, and the decision tree — Chamberlain IRB Handbook (PDF)
- Chamberlain's current DNP handbook (©2026) — the prescreening expectation and the NR-705 filing — DNP Project & Practicum Handbook (PDF)
- How the regulations define research, and a human subject — 45 CFR 46.102 (Cornell LII)
- OHRP on improvement work that is not research, on activities that are both, and on why publication plans do not settle it — Quality Improvement Activities FAQs, OHRP
- The exempt categories — 45 CFR 46.104 (Cornell LII)
OHRP's quality improvement FAQs predate the 2018 revisions and cite the earlier numbering; definitions here are given in current form. Chamberlain's current handbook or IRB portal is the authority here; where it departs from anything on this page, follow Chamberlain. This is an independent practice, unaffiliated with Chamberlain University.