| Course | NU704 Biostatistics, Epidemiology, and Social Determinants of Population Health |
|---|---|
| Unit | Unit 9 |
| Paper type | Data analysis plan |
| Length | About 980 words, 6 pages double-spaced |
| Format | APA 7 student paper |
| School | Purdue University Global |
| Program | Doctor of Nursing Practice (DNP) |
| Updated | October 2026 |
Free sample paper for NU704 Unit 9
From Discharge Navigator to Run Chart: A Unit 9 Data Analysis Plan for an Insulin Discharge Readiness Project
[Student Name]
Purdue University Global
NU704: Biostatistics, Epidemiology, and Social Determinants of Population Health
Unit 9 Data Analysis Plan
[Instructor Name]
[Date]
The project, units and data systems are composite and illustrative; the methods and sources cited are real.
Design
This quality improvement initiative compares a baseline period with an intervention period. Baseline covers the year ending the day the bundle begins. The intervention period is six months, beginning with a four-week pilot on 5 West, after which 6 West joins. Because the team expects to adjust the bundle in plan-do-study-act cycles as it goes, so analysis must show change over time rather than only compare two averages.
Measures and Definitions
The primary outcome is the 30-day glucose-related return rate. The numerator is the number of eligible patients with an emergency department visit or admission at this hospital no more than 30 days after leaving with a primary or secondary diagnosis of hypoglycemia or hyperglycemia. The denominator is all eligible patients discharged that month. Eligible patients are adults discharged from 5 West or 6 West on insulin started or changed to a pen during the stay, identified from the medication administration record and discharge medication list. Returns to other hospitals will be missed; this limitation will be reported.
Process measures are the return demonstration rate, from the new structured field in the discharge navigator; the bedside pen delivery rate, from the pharmacy delivery log; and the 72-hour call completion rate, from the diabetes educator's call log. Each uses eligible discharges as the denominator.
To make sure the bundle does not slow discharges, the plan tracks the median gap between discharge order and departure, taken from the bed management system.
If a measure cannot be defined in one sentence with a numerator and denominator, it is not ready to collect.
Data Collection and Management
An informatics analyst will run a monthly report on the first business day of the following month, listing eligible discharges with the process fields and returns within 30 days. Because returns need 30 days to occur, each month's outcome will be final 35 days after month's end. Each month the project lead pulls ten records at random and compares the report line by line with the chart. The working file lives on the hospital's secure network drive behind a password, with patient identifiers replaced by study numbers once returns are linked.
Analysis 1: Run Charts for Change Over Time
Process measures and the outcome will be plotted monthly on run charts, with the median of the baseline months drawn as a center line. Run charts suit improvement projects because they show change as it happens and can be read using a small set of probability-based rules (Perla et al., 2011). The plan will use four signals. A shift is declared when at least six consecutive months fall above it, or all below it. For a trend the rule needs five consecutive months climbing steadily, or five falling. The number of runs, meaning stretches between median crossings, is compared with the range chance would produce, and too few or too many flags a pattern. A single month far outside the rest, an astronomical point, is investigated on its own. Any point sitting exactly on the median is skipped when shifts and runs are counted.
With monthly data and only six intervention months, a shift may not be detectable for the outcome. For the process measures, which change quickly, weekly plotting during the pilot will provide enough points.
Analysis 2: Before-and-After Comparison
For the primary outcome, the plan will compare the proportion of returns in the six intervention months with the twelve baseline months. Results will be reported as the risk in each period, the relative risk with its 95 percent confidence interval, the absolute risk reduction with its interval and the number needed to treat. A significance test suited to two proportions, with an exact version if cells are sparse, will be reported with an exact p-value, but interpretation will rest on the intervals. Odds ratios will not be used for the main comparison, since the outcome is common enough that they would exaggerate the change relative to the risk ratio (Viera, 2008).
Analysis 3: Stratified Results
All measures will be reported by preferred language, by three age bands with 65 and 80 as the cut points, by literacy screen result and by food insecurity screen result. Because counts will be small, these will be presented as tables with counts and proportions rather than tested statistically, and reviewed alongside case summaries of every return in the smaller groups.
Reporting
The write-up will use SQUIRE 2.0, whose items require the setting, the bundle and every revision to it, the measures and the analytic methods to be described (Ogrinc et al., 2016). Run charts will be annotated with the dates of each plan-do-study-act change so that readers can link changes in the data to changes in the bundle.
Software
A spreadsheet template will draw each run chart, compute its median and flag the four signals automatically. The before-and-after calculations will be checked with a free online calculator and by hand.
Limitations of the Plan
Comparing an earlier year with a later half-year leaves other explanations open, and returns to other hospitals will be missed. Small monthly numbers make the outcome run chart slow to signal. Documentation of the return demonstration could be inflated by completing the field without observation, which the monthly chart audit is designed to detect.
Conclusion
The plan defines every measure precisely, sets a monthly collection rhythm with an accuracy check, uses run charts for change over time, intervals for the main comparison and stratified tables for equity, and reports according to SQUIRE 2.0. It fits a quality improvement project with modest numbers and an evolving intervention.
References
Ogrinc, G., Davies, L., Goodman, D., Batalden, P., Davidoff, F., & Stevens, D. (2016). SQUIRE 2.0 (Standards for QUality Improvement Reporting Excellence): Revised publication guidelines from a detailed consensus process. BMJ Quality & Safety, 25(12), 986-992. https://doi.org/10.1136/bmjqs-2015-004411
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
Viera, A. J. (2008). Odds ratios and risk ratios: What's the difference and why does it matter? Southern Medical Journal, 101(7), 730-734. https://doi.org/10.1097/SMJ.0b013e31817a7ee4
The Purdue Global NU704 Unit 9 instructions
Unit 9 commonly asks students to write the data analysis plan for their DNP project. Prompts usually expect the design, every measure defined with numerator, denominator and data source, the collection schedule and data management, the analysis for each measure, subgroup or equity analysis, reporting standards and limitations. Instructors reward plans precise enough that someone else could carry them out, and methods matched to the design: run charts for quality improvement projects tracking change over time, with comparisons and intervals for the main outcome. Data accuracy checks and privacy protections are often required. Sources typically include a reference on run charts, the SQUIRE 2.0 reporting guideline and a methodological source supporting the chosen effect measures. Plans should also acknowledge what the design cannot show and how the analysis will handle small numbers in subgroups.
Inside the NU704 Unit 9 example
The plan opens with the design, a baseline year and a six-month intervention with plan-do-study-act cycles. Measures are defined one by one, each with numerator, denominator and source, under a highlighted line that a measure not definable in one sentence is not ready to collect. Collection and management cover the monthly report, the 35-day lag for outcomes, the ten-record audit and secure storage. Analysis 1 uses run charts with four signals, citing Perla and colleagues, and a margin note matches plotting intervals to how fast each measure changes. Analysis 2 sets the before-and-after comparison with intervals and explains, with Viera, why odds ratios are avoided. Analysis 3 stratifies by language, age band and screens, flagged by a second note. SQUIRE 2.0 reporting, software and limitations complete the plan.
Where the points sit in the NU704 Unit 9 rubric
Data analysis plans are scored on alignment with the design, precision of measure definitions, feasibility of collection, appropriateness of analysis, equity analysis, reporting and limitations. Alignment means the plan analyzes change over time for a quality improvement project rather than forcing a trial-style test. Definition credit requires numerators, denominators and sources that leave no ambiguity. Feasibility is judged by named data systems, schedules and an accuracy check. Analysis marks reward run chart rules applied correctly and comparisons reported with effect sizes and intervals. Stratification by relevant groups, with appropriate handling of small numbers, shows equity thinking. A recognized reporting guideline and annotated charts demonstrate scholarly standards. Honest limitations, such as missed returns to other hospitals, complete the plan. Clear structure and accurate citations of the run chart reference and SQUIRE 2.0 round out the marks.
Common NU704 Unit 9 mistakes, and how to avoid them
Plans lose marks most often when measures are named but not defined, leaving the reader unsure who counts in the denominator. Another frequent error is choosing a statistical test designed for randomized trials when the project is a quality improvement initiative tracked over time. Some students forget that 30-day outcomes need time to accumulate, so monthly results are reported too early. Others plan subgroup tests on groups of a few patients, producing meaningless p-values. Omitting a balancing measure or an accuracy check leaves the plan vulnerable. Begin by writing one sentence per measure that a stranger could use to pull the same number from your hospital's systems. Then pick the display that best shows whether it changed. Your project design and measures are enough to start a custom analysis plan.
Get an NU704 Unit 9 example written to your instructions
Send the project's design, measures and the plan instructions with the rubric. A custom plan defines each measure, data source, timing and analysis method, including run charts where they fit. It comes back within 24-48h, and the first is free. The paper above is an original model document written by our desk, not a submitted student paper and not an official Purdue University Global document.
More NU704 papers and DNP samples
- NU704 Unit 1: Biostatistics in Practice Post
- NU704 Unit 2: Regression Interpretation Paper
- NU704 Unit 3: Social Determinants Analysis
- NU704 Unit 4: Statistical Test Selection Worksheet
- NU704 Unit 5: Causal Inference Paper
- NU704 Unit 6: Secondary Data Appraisal
- NU704 Unit 7: Measures of Association Problem Set
- NU704 Unit 8: Significance Interpretation Paper
- NU704 Unit 10: Statistical Literacy Reflection
- NU713 Unit 9: Structural Framework Application, WHO
- NU703 Unit 9: Practice Problem Statement, Depression
- NU733 Unit 9: Advocacy Letter to the Health Officer
- NU702 Unit 9: Project Classification Paper
NU704 Unit 9 questions, answered
What is a run chart?
A graph of a measure over time with the median marked, used in quality improvement to detect nonrandom change using defined rules.
Does a quality improvement project need statistical tests?
Often run charts are the main analysis, sometimes with a simple before-and-after comparison. The plan should match the project's design.
What belongs in a data analysis plan?
Measures and their definitions, data sources, collection schedule, data management, analysis methods and how results will be displayed.
Where can I find a free NU704 Unit 9 sample paper?
The full NU704 Unit 9 data analysis plan appears on this page. It defines every measure for an insulin discharge project, sets monthly collection with an accuracy audit and uses run charts, interval-based comparisons and stratified tables.
Why does the NU704 Unit 9 plan wait 35 days to report each month's outcome?
Returns are counted up to 30 days after discharge, so a patient discharged late in a month may return in the next. Waiting 35 days after the month ends ensures each month's outcome is complete.