| Course | DNP 955 DPI Project: Part I |
|---|---|
| Topic | Topic 3 |
| Paper type | Population and question refinement |
| Length | About 980 words, 3 pages |
| Format | APA 7 student paper |
| School | Grand Canyon University (GCU) |
| Program | DNP |
| Updated | October 2026 |
Free sample paper for DNP 955 Topic 3
One Hundred Eighty-Six Patients Who Were Never Ours: Tightening the Population and Question for a Statin Outreach Project After Checking the Extract
[Author Name]
College of Nursing and Health Care Professions, Grand Canyon University
DNP-955: DPI Project: Part I
Topic 3 Assignment
[Instructor Name]
June 18, 2029
The organization, setting and figures below are a composite written as a model document. No real employer, client, colleague or patient is described.
One Hundred Eighty-Six Patients Who Were Never Ours: Tightening the Population and Question for a Statin Outreach Project After Checking the Extract
What the Question Said and What the Extract Contained
The project's question, as drafted, concerned people between 40 and 54 whose type 2 diabetes is looked after in the composite network's primary care clinics. The baseline in Topic 1, 49 percent on a statin, came from a registry extract of 1,198 patients. Before tightening the question, I checked how the extract defined its population, because a clinical question can only be answered for the patients the data actually contain.
The extract's rule was simple: any patient aged 40 to 54, attributed to the network on the extraction date, with a diabetes diagnosis code recorded anywhere in the record in the past 24 months. Anywhere includes emergency department and urgent care encounters, outside records imported through the health information exchange and a single billing code from a laboratory visit. That rule brings in patients the network's primary care clinicians have never managed for diabetes.
Two Groups Inside One Denominator
I split the 1,198 into two groups. The first, 1,012 patients, had been seen in primary care two or more times in two years, and their diabetes was either listed as a problem or confirmed by an A1c result at or above 6.5 percent. These are the patients the project intends: people whose diabetes is managed, or should be managed, by the network's clinicians. The second, 186 patients, did not meet that definition. On review of 40 of them, most had been seen once in the network's emergency department or urgent care, had a diabetes code carried over from an outside record, and had no primary care visit in two years; several had moved away, and a few had type 2 diabetes recorded in error.
The two groups differ by a substantial margin. In the managed group, 553 of 1,012, or 55 percent, had a statin ordered or dispensed in the past 12 months. In the other group, 33 of 186, or 18 percent, did. The original baseline of 49 percent was an average of two populations, one of which the project cannot reach through primary care outreach because those patients have no primary care relationship to use. Including them would make the baseline look worse than it is and would guarantee that the project could never close the gap it claimed to address.
The Rebuilt Extract and the Corrected Baseline
The extract has been rebuilt to match the intended population: attributed adults aged 40 to 54 who have made two or more primary care visits in the network within two years, whose type 2 diabetes appears as a listed problem or is confirmed by an A1c at or above 6.5 percent, and who have no recorded diagnosis of atherosclerotic cardiovascular disease, the group the diabetes standards say should be offered a statin of at least moderate intensity (American Diabetes Association Professional Practice Committee, 2025). Patients with a documented contraindication, such as pregnancy or a prior statin-related adverse event, are removed from the denominator, which removes 41 more. The corrected baseline is 553 of 971, or 57 percent.
Both figures will be reported in the proposal, side by side: the original 49 percent from the broad extract, and the corrected 57 percent from the rebuilt one, with a sentence explaining the difference. A committee member who saw 49 percent in an earlier draft should not have to wonder where it went. The 186 patients outside the managed population are not a problem to ignore; they are a different problem, about connecting people to primary care, that this project will note and not attempt.
The Question, Tightened
The question is tightened in the same pass, with each element specified to the level the data can support (Stillwell et al., 2010). Population: the managed group defined above, ages 40 through 54, without atherosclerotic cardiovascular disease or a documented contraindication to statins. Intervention: nurse care manager outreach using a shared decision aid, with a moderate-intensity statin order pended for prescriber signature under protocol when the patient chooses to start. Comparison: the same population during the six months before implementation. Outcome: the proportion with a statin ordered or dispensed in the prior 12 months, using the network's measure definition. Time: measured monthly from March 2029 through February 2030, with implementation from September 2029.
The outcome is stated as ordered or dispensed, not as adherence or lower cholesterol, because that is the level the data support. Pharmacy fill data cover roughly three in five of these patients, and LDL cholesterol is measured irregularly in this age group, so a question promising reduced LDL or sustained adherence would promise results the project could not measure. The written question now reads: in attributed adults aged 40 to 54 with primary care-managed type 2 diabetes and no atherosclerotic disease or statin contraindication, does nurse care manager outreach with a shared decision aid, compared with the preceding six months, increase the monthly proportion with a statin ordered or dispensed in the prior 12 months, between September 2029 and February 2030?
Why This Was Worth Doing Now
Tightening the population raised the baseline from 49 to 57 percent, which makes the gap smaller and the project look less dramatic. It also makes the project honest. The outreach list for the care managers will now contain people who have a primary care relationship to call on, the decision aid can be used with patients who have been seen and can be seen again, and the evaluation will compare like with like. The decision aid was tested with adults who had type 2 diabetes and an ongoing clinical relationship, although in specialty diabetes consultations rather than primary care, which is another reason to keep the population limited to patients with a clinician who knows them (Weymiller et al., 2007).
References
American Diabetes Association Professional Practice Committee. (2025). 10. Cardiovascular disease and risk management: Standards of care in diabetes-2025. Diabetes Care, 48(Suppl. 1), S207-S238. https://doi.org/10.2337/dc25-S010
Stillwell, S. B., Fineout-Overholt, E., Melnyk, B. M., & Williamson, K. M. (2010). Evidence-based practice, step by step: Asking the clinical question: A key step in evidence-based practice. American Journal of Nursing, 110(3), 58-61. https://doi.org/10.1097/01.NAJ.0000368959.11129.79
Weymiller, A. J., Montori, V. M., Jones, L. A., Gafni, A., Guyatt, G. H., Bryant, S. C., Christianson, T. J. H., Mullan, R. J., & Smith, S. A. (2007). Helping patients with type 2 diabetes mellitus make treatment decisions: Statin choice randomized trial. Archives of Internal Medicine, 167(10), 1076-1082. https://doi.org/10.1001/archinte.167.10.1076
The DNP-955 Topic 3 assignment instructions
Topic 3 in DNP-955 asks students to refine their project population and question. Strong papers compare the question as written with what the data extract actually contains, since the two often differ, and identify groups inside the denominator that the intervention could never reach. Students should rebuild the extract to match the intended population, report the corrected baseline alongside the original and tighten the question element by element to the level the data can support. The paper should explain why making the project look less dramatic, for example by raising the baseline, makes it more honest and protects the evaluation later. Counts should be exact.
How the DNP-955 Topic 3 example is put together
The finished example catches something. The question specifies adults with a condition managed in primary care, and the extract that produced the baseline includes anybody with the diagnostic code recorded anywhere, which brings in patients seen once in an emergency department and never again. The two groups differ by a substantial margin, and the baseline rate differs with them. The paper works out which group the project actually intends and then rebuilds the extract to match, reporting the corrected baseline alongside the original. The question is tightened in the same pass, with the timeframe made explicit and the outcome specified to the level the data can actually support. Both population figures appear in the paper with the rule that produced each of them.
Reading the DNP 955 Topic 3 rubric
Faculty reward a population checked against the extract. An Excellent paper splits 1,198 patients into 1,012 managed in primary care and 186 with a single visit or an unconfirmed diagnosis. Credit goes to rebuilding the extract with attribution and two visits in two years, raising the baseline from 49 to 57 percent. The rubric values a question tightened with Stillwell and colleagues' PICOT elements and ADA criteria. Strong papers report both baselines and why they differ. Top marks go to arguing that a smaller gap makes the project honest. Both baselines are shown.
DNP 955 Topic 3 help: mistakes that cost marks
Tightening a question without checking it against the extract is exactly what this topic sets out to surface, and the mismatch otherwise appears during implementation. A second weakness is choosing the population that is easiest to extract rather than the one the intervention can reach, which produces a project measuring people it cannot affect. Marks also go for replacing a baseline without showing the original, since the change itself is a finding. Questions specifying outcomes the data cannot register at that resolution promise measurement nobody can perform. Timeframes left implicit leave the evaluation window undecided. Populations described without a selection rule cannot be reproduced. Populations that widen between the question and the analysis dilute any effect the project produces.
Get a DNP-955 Topic 3 example written to your instructions
Send the DNP-955 Topic 3 instructions and the rubric your classroom posts, with your question and extract details. We write a custom example to those criteria, comparing the specified population against the extracted one, deciding on reach and reporting the corrected baseline, in 24 to 48 hours. The first is free.
More DNP 955 papers and DNP samples
- DNP 955 Topic 1: Practice Problem Revisit
- DNP 955 Topic 2: Project Decision Record
- DNP 955 Topic 4: Literature Arranged by Argument
- DNP 955 Topic 5: Framework and Design Defense
- DNP 955 Topic 6: Site Approval DQ Pair
- DNP 955 Topic 7: Measurement and Analysis Plan Draft
- DNP 955 Topic 8: Proposal Assembly and Consistency Pass
- DNP 960A Topic 3: Fidelity Monitoring Plan
- DNP 840 Topic 3: Interprofessional Team Analysis
- DNP 965A Topic 3: Bounded Interpretation Section
- DNP 838A Topic 3: Faculty Staffing Plan
DNP 955 Topic 3 questions, answered
How do the two populations end up different?
Through the extraction rule, usually. A question about patients managed in primary care is precise; an extract selecting anybody with the diagnostic code recorded anywhere is broader and much easier to run. Nobody decides to do this, and it happens constantly, because the extract is built by someone answering a slightly different question from the one in the proposal.
Which population should win?
The one your intervention can reach. If the change operates in primary care, patients seen once in an emergency department cannot benefit from it and should not be in the denominator diluting the result. Choosing on reach rather than on ease of extraction is the judgment this topic is testing.
Do I report both baselines?
Yes. The original figure appears in earlier documents and somebody will notice if it changes silently. Showing both, with the reason for the correction, demonstrates that you found a real problem and fixed it. Replacing one number with another and saying nothing is the version that raises questions at defense.
Where can I find a free DNP 955 Topic 3 sample paper?
The refinement paper for DNP 955 Topic 3 covers what the question said and what the extract contained, two groups in one denominator, the rebuilt extract and corrected baseline, the question tightened and why it was worth doing now. Post your DNP 955 Topic 3 prompt and get a free paper.
Should both baselines be reported in DNP 955 Topic 3?
Yes. The original 49 percent appears in earlier documents, so the paper reports it beside the corrected 57 percent and explains which patients account for the difference.