| Course | NR 714 Application of Analytic Methods II |
|---|---|
| Week | 5 |
| Paper type | Data analysis plan |
| Length | About 1,140 words, 5 pages |
| Format | APA 7 student paper |
| School | Chamberlain University |
| Program | Doctor of Nursing Practice (DNP) |
| Updated | October 2026 |
Free sample paper for NR 714 Week 5
Data Analysis Plan: Comparing Peripheral IV Catheter Failure Before and After a Securement Bundle on 4 North, Fixed Before the Data Arrive
Student Name
College of Nursing, Chamberlain University
NR 714: Application of Analytic Methods II
Instructor Name
Month Day, Year
Data Analysis Plan: Comparing Peripheral IV Catheter Failure Before and After a Securement Bundle on 4 North, Fixed Before the Data Arrive
Purpose of Fixing the Plan Now
This plan was approved by the unit's quality council on 30 August 2026, before any post-bundle record was extracted. Every decision below is committed in advance: the comparison, the procedure, the handling of incomplete records, the subgroup and the result that would count as no change. Writing these down first prevents the analysis from being shaped by the results, which is easy to do without noticing once the numbers are on the screen (Polit & Beck, 2021).
The Practice Question, Restated
Among adult patients' peripheral intravenous catheters on the 36-bed medicine unit 4 North, is the rate of premature catheter failure per 1,000 catheter-days lower in the six months after the securement bundle began than in the six months before it?
Definitions Carried Forward Unchanged
The outcome, population, exclusions and extraction are exactly those set out in the outcome measure write-up. Failure is removal before the end of therapy for any of the five failure reasons (infiltration or extravasation, occlusion, dislodgement, phlebitis, leakage), read from the removal reason field. Catheter-days start once the catheter is both in place and on the unit, and stop at whichever comes first of removal, transfer or discharge. Catheters removed within one hour of arrival, midlines, central lines and catheters in comfort-care patients are excluded. No definition will be altered for the post period. If the record system changes a field during the project, the change and its date will be reported, and the old mapping will be applied to the new field wherever that is possible.
Periods Compared
Baseline: 1 January to 30 June 2026, the period already reported, with 437 failures over 4,236 catheter-days. Post period: 1 September 2026 to 28 February 2027, also 181 days. July and August 2026 are plotted on the run chart but excluded from the comparison, because July was not part of the baseline extraction the council approved and bundle training began on 12 August, which makes August neither before nor after.
Expected Records
At the baseline pace, the post period should produce about 1,400 catheters and 4,200 catheter-days. If fewer than 1,260 analyzable catheters are obtained, 10% below that, the shortfall will be explained before any comparison is reported.
The Primary Comparison and Why It Fits
The outcome is a count of events over a measured exposure, so the primary analysis compares incidence densities. The estimate is the rate ratio, post-period rate divided by baseline rate, with a 95% confidence interval computed on the log scale from the two event counts. A Poisson approach fits because each catheter-day is an exposure unit and failures are counted over them; a comparison of proportions would ignore the possibility that the bundle lengthens dwell time and so changes the exposure. Analysis will be run in R, version 4.4, by the unit's data analyst, with the project lead checking every figure by hand from the counts (Celentano & Szklo, 2019).
Secondary Comparisons, Named Now
Three secondary comparisons are fixed in advance, and no others will be reported as findings. First, the percentage of catheters that fail, before and after, as a risk difference with a 95% confidence interval. Second, the dislodgement rate ratio, because dislodgement is the reason securement most directly addresses and it has enough events, 96 at baseline, to be estimated. Third, a monthly u-chart of failures per 1,000 catheter-days, with center line and limits from January to August 2026, read with the standard shift and trend rules. Infiltration, occlusion, phlebitis and leakage rates will be reported as description only.
One subgroup is planned: catheters inserted on the unit against catheters that arrived from the emergency department, because the bundle is applied at insertion and arriving catheters receive it only at their first dressing change. No other subgroup will be examined as a finding, however interesting it looks once the data are in.
Incomplete and Absent Records
A blank removal reason will be reviewed against the nursing notes by the project lead and a second nurse. If both agree the notes describe a failure, it is counted as one. If the reason is still unknown, the primary analysis counts the catheter as not failed, and a sensitivity analysis counts it as failed; both results will be reported. A catheter with neither an insertion time nor an arrival time cannot contribute catheter-days and will be excluded and counted. If such exclusions exceed 3% of catheters, the excluded group will be compared with the analyzed group on age, arrival route and time of arrival. Multiple imputation was considered and rejected for the primary analysis, because the missing piece is the exposure itself rather than a covariate, and imputing exposure would add more assumption than it removes (Sterne et al., 2009).
Patients who move off the unit stop contributing catheter-days at transfer, and a failure after transfer is not counted. A patient readmitted to 4 North in the period contributes again with each new catheter.
The Smallest Detectable Difference
With about 437 failures expected at baseline and about 4,200 catheter-days after, a two-sided test at the 5% level has roughly 80% power to detect a rate ratio near 0.82, a fall from 103.2 to about 84.6 failures per 1,000 catheter-days. A smaller true effect, such as a 10% reduction, could easily produce a confidence interval that includes 1.0. The unit has accepted this limit in advance: the project is sized to detect a reduction of about one failure in five, which is the size of change that would justify the bundle's supply cost.
What Would Count as No Change
The result will be called a reduction if the upper limit of the 95% confidence interval for the rate ratio is below 1.0. It will be called no change shown if the interval includes 1.0, and the interval will be reported in full so that a reader can see whether a meaningful reduction remains possible. If the interval lies entirely above 1.0, the bundle will be described as associated with more failures and the unit will stop it. A recommendation to extend the bundle to other units requires both a reduction and bundle use on at least 80% of audited catheters, since a fall in failures without the bundle being applied would point to something else.
Balancing Measures
Two balancing measures will be reported with the outcome: medical adhesive-related skin injuries recorded in the incident system, and the added supply cost per catheter, currently $3.10.
References
Celentano, D. D., & Szklo, M. (2019). Gordis epidemiology (6th ed.). Elsevier.
Polit, D. F., & Beck, C. T. (2021). Nursing research: Generating and assessing evidence for nursing practice (11th ed.). Wolters Kluwer.
Sterne, J. A. C., White, I. R., Carlin, J. B., Spratt, M., Royston, P., Kenward, M. G., Wood, A. M., & Carpenter, J. R. (2009). Multiple imputation for missing data in epidemiological and clinical research: Potential and pitfalls. BMJ, 338, b2393. https://doi.org/10.1136/bmj.b2393
NR 714 Week 5 instructions, in plain terms
Restate the practice question exactly as it will be analyzed, and carry the outcome definition forward from your earlier write-up without changing a word. Name the periods or groups compared, and give the reason for any time left out. Choose the procedure from the measurement level and the design, and say why it fits. List the records you expect, with a floor below which you will explain the shortfall before reporting. Fix the primary comparison, name every secondary comparison and subgroup now and say that no others will be treated as findings. Write the rules for incomplete or absent records and for patients who move into or out of the group. State the smallest difference the project could realistically detect, and define in advance the result that will count as no change, along with any balancing measures.
Inside the NR 714 Week 5 example
The sample opens by noting the plan was approved before any post-bundle record was pulled. It restates the question, carries the definition forward unchanged and sets two periods of 181 days. The expected-records section names about 1,400 catheters with a floor of 1,260. The primary comparison is a rate ratio from event counts, chosen over a comparison of proportions because the bundle may lengthen dwell. Secondary comparisons are a risk difference, the dislodgement rate and a monthly u-chart, with one subgroup by insertion location. Missing reasons go through a two-nurse review, then primary and sensitivity counts, and imputation is rejected with a reason. A power estimate gives a detectable rate ratio near 0.82, and the decision section ties a recommendation to both a reduction and at least 80% bundle use.
Where the points sit in the NR 714 Week 5 rubric
The costliest failure is a plan with nothing committed in advance, where every real decision waits for the data. Next is a procedure that does not fit the design, picked from habit instead of from the measurement level and the comparison. Marks also go when absent data are never mentioned, when exclusion rules are missing so the sample can quietly change shape and when subgroups are left open-ended. Sample size treated as whatever turns up, with no sense of what it could detect, costs points, as does an outcome defined differently than it was two weeks earlier. Plans score well when they write down the result that would count as no change, because that is the commitment that keeps the later write-up honest.
NR 714 Week 5 help from the desk
Writing down decisions before you have data feels premature, which is why most draft plans leave the hard choices open. Tell us your outcome, design and expected numbers, and we can help you choose and justify the procedure, write rules for missing records, limit subgroups, estimate the smallest detectable difference and define what no change would look like. We can also help you check that your definitions match your earlier assignments. The plan reflects your project and your own decisions about what to commit to.
Get a NR 714 Week 5 example written to your instructions
Send the instructions, the rubric and a description of your project: the setting, the outcome, the window you have and roughly how many records you expect. An original plan is written to those constraints and returned inside 24 to 48 hours. The first example costs nothing and is written only for you.
More NR 714 papers and DNP samples
- NR 714 Week 1: Practice Question Post
- NR 714 Week 2: Outcome Measure Write-Up
- NR 714 Week 3: Epidemiologic Rates Write-Up
- NR 714 Week 4: Paired Analysis Write-Up
- NR 714 Week 6: Data Analysis Write-up
- NR 714 Week 7: Evaluation Plan
- NR 714 Week 8: Results Interpretation Presentation
- NR 709A Week 5: Handover Packet Outline
- NR 707A Week 5: Missing Data Memo
- NR 705A Week 5: Measure Definitions
- NR 702A Week 5: Aim Statement Draft
NR 714 Week 5 questions, answered
How does this differ from the evaluation plan later in the course?
The analysis plan settles what will be done with numbers once they exist. The evaluation plan is the wider document around it, covering what gets measured, by whom, on what schedule, and how the practice change as a whole will be judged. Sections order the two differently, but the analysis plan is the narrower and more technical, and it usually sits inside the other.
Should the plan include a power calculation?
Some sections require one and many ask instead for a reasoned statement about sample size. Either way the honest content is the same: how many records the project window will realistically produce, and how large a difference that number could detect. Writing that the sample will be adequate without ever saying adequate for what is the version that loses marks.
Can the plan change once collection starts?
It can, and the plan is stronger for saying how. A short passage stating that departures will be documented and reported with their reasons is ordinary practice, and it distinguishes a considered change from an analysis quietly rerun until something reached significance. The following week's write-up is where those departures get declared, so the two documents are built to fit together.
Where can I find a free NR 714 Week 5 sample paper?
A full data analysis plan comparing peripheral IV catheter failure before and after a securement bundle, fixed in advance, is on this page.
What does the NR 714 Week 5 plan say would count as no change?
A 95% confidence interval for the rate ratio that includes 1.0, reported in full so readers can see whether a meaningful fall is still possible.