| Course | DNP 850B Project Proposal |
|---|---|
| Module | Module 6 |
| Paper type | Data analysis plan |
| Length | About 1,020 words, 6 pages |
| Format | APA 7 student paper |
| School | Aspen University |
| Program | DNP |
| Updated | September 2026 |
Free sample paper for DNP 850B Module 6
Answering the Question the Project Asked: A Data Analysis Plan for Nurse Titration of Home Readings
Student Name
Doctor of Nursing Practice Program, Aspen University
DNP 850B: Project Proposal
Instructor Name
Month Day, Year
Answering the Question the Project Asked: A Data Analysis Plan for Nurse Titration of Home Readings
An analysis plan written before the data arrive protects a project from choosing the tests that make its results look best. This paper sets out that plan for my DNP project, a 12-week nurse-led program of home readings and protocol titration in a health center clinic. The project's question asks whether, among clinic patients aged 18 to 75 with uncontrolled hypertension, the program lowers mean office systolic pressure and raises the proportion under 140/90 mm Hg by week 12, relative to the previous year's usual care. The plan matches a statistic to each part of that question, states the assumptions and what happens if they fail, and explains how time-ordered process data and missing values will be handled.
Matching the Analysis to the Question
Each element of the question is paired with a method in the table. The plan was reviewed with the committee's quality improvement faculty member before the proposal defense, so that no test is added after the data are seen.
| Question element | Measure | Analysis |
|---|---|---|
| Change in office systolic pressure among enrolled patients | Week-12 mean minus baseline mean | Paired t-test; Wilcoxon signed-rank if assumptions fail; mean change with 95% CI |
| Share below 140/90 at week 12 | Proportion controlled | Proportion with 95% CI |
| Comparison with usual care in the prior year | Change in systolic pressure; proportion controlled | Linear regression of week-12 systolic pressure on group, adjusted for baseline; difference in proportions with 95% CI |
| How the program was delivered | Weekly transmission, time to action, titrations | Run charts with median and run rules |
| Safety | Balancing events | Counts and rates per 100 patient-weeks |
| Adherence behaviors | Hill-Bone total score | Paired t-test on baseline and week-12 scores |
Descriptive Statistics
The sample will be described first: age, sex, preferred language, insurance, number of blood pressure medications at baseline and baseline systolic and diastolic pressures. Continuous variables will be reported as means and standard deviations, or medians and interquartile ranges if skewed, and categorical variables as counts and percentages. The same description will be produced for the historical comparison group, placed side by side, so that differences between the groups are visible before any comparison of outcomes.
Primary Analysis
The primary analysis compares each enrolled patient's baseline and week-12 office systolic pressure using a paired t-test, with the mean change reported with its 95% confidence interval. The test assumes that the differences between paired readings are approximately normal. That assumption will be checked with a histogram and a normal probability plot of the differences; if they show marked skew or outliers that cannot be explained, the analysis switches to the Wilcoxon signed-rank test and reports the median change. The significance level is set at .05, two-sided. Because baseline control is zero by definition, since every enrolled patient was uncontrolled, the proportion controlled at week 12 will be reported with a 95% confidence interval rather than tested against baseline.
Comparison With Usual Care
For the comparison with the historical group, the plan uses linear regression with week-12 systolic pressure as the outcome and group and baseline systolic pressure as predictors. Adjusting for baseline in this way is more efficient than comparing change scores and handles imbalance in starting values better (Vickers & Altman, 2001). Age and preferred language will be added in a second model if the groups differ on them. As noted in the sample section, this comparison is underpowered, so the adjusted difference will be reported with its confidence interval and interpreted as an estimate rather than a test.
Regression to the Mean
Patients are enrolled because their blood pressure was high, and a group selected on high values tends to show lower values later even without intervention (Barnett et al., 2005). The design limits this by requiring two elevated readings on separate days and averaging them for baseline. The analysis limits it further: the historical group was selected by the same rule and will show the same regression, so the adjusted comparison removes much of it. Chapter four will state that the paired within-group change includes some regression and should not be read alone as the program's effect.
Process Measures Over Time
Each weekly process measure will get its own run chart, centered on the median of weeks one to four. Run chart rules will be applied to decide whether a change is a signal or noise: a run of six or more points that all sit above the median or all sit below it, five or more points in a row all rising or all falling, a count of runs outside the expected range, or one point far out of line with the others (Perla et al., 2011). The charts will be reviewed at team huddles in weeks 4, 8 and 12 to guide changes in workflow, and each change will be annotated on the chart.
Missing Data
Patients without a week-12 office reading will be counted and described. The primary analysis will use complete cases, and a sensitivity analysis will carry each missing patient's last office reading forward, so that the effect of missing data on the result can be seen. If more than 15% of patients are missing, the patients with and without follow-up will be compared on baseline characteristics.
Statistical and Clinical Significance
A statistically significant change may still be too small to matter. The project will treat a mean fall of 5 mm Hg or more in systolic pressure as clinically meaningful, a figure consistent with the effects seen in the trials reviewed in chapter two. Results will be interpreted against that threshold as well as the p value. Analyses will be run in a standard statistical package, and the code and data dictionary will be kept so a committee member can reproduce every figure.
Conclusion
The plan pairs each part of the project question with a method chosen in advance, checks assumptions, adjusts the comparison for baseline, accounts for regression to the mean, reads process data over time, handles missing values openly and judges results against a clinical threshold. Writing it now fixes the rules before the data can influence them.
References
Barnett, A. G., van der Pols, J. C., & Dobson, A. J. (2005). Regression to the mean: What it is and how to deal with it. International Journal of Epidemiology, 34(1), 215-220. https://doi.org/10.1093/ije/dyh299
Perla, R. J., Provost, L. P., & Murray, S. K. (2011). The run chart: A simple analytical tool for learning from variation in healthcare processes. BMJ Quality & Safety, 20(1), 46-51. https://doi.org/10.1136/bmjqs.2009.037895
Vickers, A. J., & Altman, D. G. (2001). Analysing controlled trials with baseline and follow up measurements. BMJ, 323(7321), 1123-1124. https://doi.org/10.1136/bmj.323.7321.1123
DNP 850B Module 6 instructions, in plain terms
In this module the proposal commits to how the data will be analyzed. With Aspen's prompt available only in the classroom, the sample follows the catalog summary that places the project's methodology in DNP 850B. An analysis plan assignment usually asks you to match each outcome to a statistical test, state assumptions and what you will do if they fail, list the descriptive statistics you will report, and set out the standard you will use for significance. Your chair may require the software you will use, a table linking the PICOT to the analysis, or a plan for missing data. Some programs involve a statistician at this point. Confirm length and source requirements, and make sure every test you list fits the level of measurement of its variable.
How this DNP 850B Module 6 example is built
At about 1,030 words, the example has ten sections. The introduction explains why the plan is written in advance and restates the PICOT question. A table matches six elements of the question to measures and analyses. Descriptive statistics come next, with the historical group described side by side. The primary analysis explains the paired t-test, the normality check and the nonparametric alternative, and why control is reported as a proportion. The comparison section explains regression adjusted for baseline. A section on regression to the mean shows how the design and analysis limit it. Run chart rules are set out for process data. Missing data are handled with complete cases and a sensitivity analysis, and a final section defines clinical significance before the conclusion.
DNP 850B Module 6 rubric: what earns full marks
Analysis plans are graded on the match between question and test, the handling of assumptions and the honesty of interpretation. The table meets the first criterion in one view, and the margin notes explain why that match is the core of the plan. Assumptions are addressed with a named check and a fallback test. Honest interpretation shows in three places: reporting the underpowered comparison as an estimate, treating the within-group change as partly regression to the mean, and naming a clinical threshold in advance. Run chart rules show knowledge of improvement methods, which DNP rubrics often value. The final marks go to APA format and to citations for the analytic choices, such as the recommendation to adjust for baseline.
DNP 850B Module 6 help from the desk
The most common error is naming tests without linking them to the question, so the committee cannot tell which result answers what. Build a table. Students also choose tests that do not fit their data, such as a chi-square test for a continuous outcome. Check the level of measurement. Another mistake is ignoring assumptions and never saying what happens if they fail. Name the check and the alternative. Papers often treat a significant p value as proof of benefit. Add a clinical threshold. Missing data are frequently left out entirely; say how you will handle and report them. Finally, do not plan tests you cannot power. Report underpowered comparisons descriptively and say so.
Write yours, or have the desk draft it
This paper is an original model document written by our desk, not a submitted student paper and not an official Aspen University document. Read it for the moves, then write your own to the instructions in your classroom. If you want one built to your exact prompt and rubric, the first custom sample is free and arrives in 24 to 48 hours.
More DNP 850B and DNP sample papers
- DNP 850B Module 1: Literature Synthesis for the Intervention
- DNP 850B Module 2: Conceptual Framework for the Project
- DNP 850B Module 3: Setting, Population and Sample
- DNP 850B Module 4: The Intervention and Its Timeline
- DNP 850B Module 5: Measures and Instruments
- DNP 850B Module 7: Ethics, Human Subjects Training and Review
- DNP 850B Module 8: Final Proposal: Literature and Methods
- DNP845 Module 1: Ways of Knowing Paper
- DNP 830 Module 1: Measuring Global Burden
- DNP 805 Module 5: System Quality Initiative With a Driver Diagram
- DNP 851B Module 5: Answering the PICOT Question
DNP 850B Module 6 questions, answered
What does DNP 850B Module 6 usually ask for?
Aspen's DNP 850B description covers solidifying the project's methodology, so a data analysis plan matched to the PICOT question is a typical assignment. Check your classroom for the prompt.
Which test fits a DNP pre-post project?
For a continuous outcome measured twice in the same patients, a paired t-test is standard, with the Wilcoxon signed-rank test if the differences are not roughly normal. Proportions are usually reported with confidence intervals or compared with McNemar's test.
What is the difference between statistical and clinical significance?
Statistical significance says a change is unlikely to be chance. Clinical significance asks whether it is large enough to matter to patients. A plan should name a clinical threshold in advance, such as 5 mm Hg.
Where can I find a free DNP 850B Module 6 sample paper?
This page carries the complete data analysis plan with its question-to-analysis table and notes in the margin, and it is free to read. The Module 5 sample defines the measures these analyses use, since the whole course follows one project.
Which statistical test fits DNP 850B Module 6?
It depends on the outcome and design. A paired t-test suits a continuous outcome measured twice in the same patients, regression suits comparisons adjusted for baseline, and run charts suit process data over time. Match each test to a part of your PICOT question.