Your WGU capstone is probably not human-subjects research. You still have to file.
Read WGU's own program guidebooks and the shape of the capstone is unmistakable: a change implemented inside one organisation, measured against indicators that organisation chose. Under the federal definition, that is usually not research at all. What it is not is self-certifying — the office says so in writing, or nobody has said it.
Most WGU capstones are single-site improvement work, which is not human-subjects research under 45 CFR 46. The landing is common and legitimate — but it arrives as a written determination from WGU's IRB office, obtained by filing, not by deciding.
What shape does WGU actually build into the capstone?
An improvement project, explicitly. WGU's published guidebook for the MSN in nursing leadership and management describes a healthcare improvement project carried through the four phases of a project lifecycle, with the candidate assessing organisational readiness with internal and external stakeholders, implementing a training plan for staff and managers on site, and then evaluating results against the key indicators chosen while planning. The competency language is worth sitting with: an improvement project made of systematic activities organised and implemented to improve organisational quality, safety and performance.
Notice what that language holds and what it omits. It holds systematic. It omits any aim of producing knowledge for anyone outside the organisation. Together, those two facts settle most WGU capstones' IRB branch before a form is opened.
The pattern repeats across the health tracks. WGU's nursing education guidebook frames its capstone as an evidence-based curriculum proposal, built on a field experience and carried into planning for a proposed curriculum change at the candidate's own programme. At baccalaureate level WGU lists, as a competency in its own right, telling improvement processes, evidence-based practice and research apart. The university teaches that distinction because it governs what you file.
What does the regulation count as research?
One sentence at 45 CFR 46.102(l) carries it. Research is a systematic investigation — the regulation names development, testing and evaluation among the forms it takes — whose design reaches for knowledge that is meant to generalize. The load sits on that last clause. Being organised, being measured, using an instrument, running a before-and-after comparison: none of it makes an activity research. Designing the answer to travel does.
Four categories of activity are then placed outside the definition outright by the same section: journalistic and scholarly work aimed at the very individuals it describes; surveillance ordered or authorised by a public health authority; records work done for criminal justice purposes; and certain authorised intelligence operations. A WGU capstone is rarely any of them, which is why the argument here almost always turns on intent rather than on a listed exclusion.
A second definition must also be met. A human subject, at 46.102(e), is a living person an investigator interacts or intervenes with, or whose identifiable private information the project obtains, uses or generates. Fail either definition and part 46 does not reach the project. See the fork that splits the tree in half.
How does OHRP read a quality improvement project?
In a fixed order. The federal Office for Human Research Protections publishes quality improvement FAQs that ask, in sequence: is it research; if so, are human subjects involved; if so, does an exemption apply; and is it covered by an assurance. Most capstones stop at the first question.
OHRP then answers the case that matches a WGU capstone almost exactly. Where the whole of an activity is putting an accepted practice into use so care improves, plus gathering data from patients or providers about how that rollout went for clinical, administrative or practical ends, the definition of research is not met — so those regulations require neither IRB review nor consent from patients or providers. The published examples are ordinary improvement work: a clinic auditing whether a widely accepted assessment is reaching the right patients, a hospital group checking whether a known procedure cut prescription errors as expected.
Its counter-example is where capstones cross the line. Bringing in an intervention nobody has tested, partly so that scientific evidence of how well it performs can be established, may be non-exempt human-subjects research even while it also improves care. If your project's novelty is the point of it, read that one twice.
Does planning to publish turn it into research?
No — and OHRP says so plainly. Publishing intent, on its own, is not sufficient to decide whether an improvement activity involves research; accounts of non-research work are published constantly, because others may want to learn from them. The same FAQ shuts the trapdoor on the other side too: a project can involve research where nobody intends to publish a line.
What governs is design intent, fixed when the project is designed. Which is why the intent paragraph in a WGU application deserves more drafting care than any other, and why we never let it be written last.
How do the two readings differ on the page?
Not by topic. Falls, sepsis screening, handoff communication and onboarding all appear on both sides of this line. The difference shows up in five design choices a reviewer can see.
| The choice | Reads as improvement at one site | Reads as research |
|---|---|---|
| Why the project exists | A named problem on a named unit, raised by that unit | A gap in the literature the project is designed to address |
| What is being implemented | A practice already accepted, applied locally | Something untested, applied to see whether it works |
| The comparison | The site against its own prior performance | Arms, controls or groups built to support inference |
| The instruments | The organisation's existing measures and audit tools | Validated instruments chosen so results transfer |
| Who the answer is for | The site's leadership, to decide what to keep | Readers elsewhere, deciding whether to adopt |
A test that costs nothing: would the project still be worth doing if nobody outside your organisation ever read a word about it? A confident yes puts you on the improvement side. Hesitation is information, and it is cheaper now than after collection.
If it is not research, why file anything at all?
Because the university reserves that call for its own office. WGU's Policy Handbook states that proposals to conduct research using human subjects are submitted to the IRB so the IRB can determine whether full review and approval are required, and it lists the screen a proposal has to pass to be deemed exempt from IRB oversight: no human subject data, no research purpose, no atypical methodology, no vulnerable populations or sensitive topics, and no risk to participants. That is a description of a decision being made by a reader — not a checklist you tick for yourself.
Published WGU capstone guidance points the same way, describing a letter of determination that the project is exempt for IRB purposes, obtained before data collection begins. Whatever the current form is called and wherever it is routed — that is set by WGU's current handbook or IRB portal, which sits behind university sign-on — the pattern holds: you file something short and precise, and the office answers in writing. The order of the walk shows where that filing sits relative to everything else.
What is the price of the wrong call, either way?
Both errors are expensive, and neither is the safe one.
- Research dressed as improvement. You collect under a claim the office never examined. The branch you actually walked was never reviewed, participants were never protected by the review that should have applied, and the data can be unusable at the point someone reads the design honestly. This is the failure that ends capstones.
- Improvement dressed as research. You volunteer a local project into a review it never needed. Now you owe a consent file, a recruitment package, a data-security plan and a site letter for a project that required a determination request — and every one of those documents creates a new surface for the board to return notes on.
- The quiet third error. An intent paragraph claiming improvement while the appendices describe research — a scale chosen because it travels, a comparison arm, a sampling frame. When claim and file disagree, the file wins.
How do you make the claim hold across the folder?
By writing every page as though someone will check it against the intent paragraph, because someone will. Use the measures the organisation already keeps. Name the unit, the process and the target in the aim. Let recruitment wording promise nothing the consent information withholds. And keep whatever you claimed about identifiers true through to the last appendix — the point where most otherwise honest improvement claims quietly collapse.
Where a design genuinely sits on the line, the fix is often small and truthful: dropping a comparison arm nobody needed, using the site's audit tool rather than a published scale, narrowing a question to the unit it is really about. Have that conversation before drafting. Where each branch ends covers what the heavier landings require when redesign is not the right answer.
What to do next
Write one paragraph describing the capstone to your site's leadership: problem, change, measure, who the answer serves. Then ask whether any part of it only makes sense as evidence for other organisations. That paragraph is the determination request in embryo.
Send it to us and one of the three consultants replies with a reading — free, no obligation. The walk, station by station shows which parts we carry, and the free application review is where a clean draft gets told it is clean.
Sources
- WGU Policy Handbook — Institutional Review Board (IRB), cm.wgu.edu
- WGU Program Guidebook — RN to MSN, Nursing Leadership and Management (PDF), wgu.edu
- WGU Program Guidebook — BSN to MSN, Nursing Education (PDF), wgu.edu
- WGU Program Guidebook — Bachelor of Science, Nursing (PDF), wgu.edu
- OHRP — Quality Improvement Activities FAQs, hhs.gov
- 45 CFR part 46 (the Common Rule), full regulatory text, govinfo.gov
Have the classification read before you draft, not after.
Send the aim statement and whatever else exists. Back comes a written reading of the landing your capstone honestly reaches, and what that landing's folder needs.
Request the free application review- whether the design reads as improvement or as research
- whether a small truthful change moves the branch
- whether you can reasonably file the determination yourself