| Course | NURS FPX 6424 Data Mining to Advance Healthcare |
|---|---|
| Assessment | Assessment 2 |
| Paper type | Data-scoped clinical question |
| Length | About 1,080 words, 6 pages |
| Format | APA 7 student paper |
| School | Capella University |
| Program | MSN |
| Updated | September 2026 |
Free sample paper for NURS FPX 6424 Assessment 2
A Clinical Question Scoped to Its Data: Predicting 30-Day Readmission After a COPD Exacerbation From Electronic Health Record Data
Student Name
Capella University
NURS FPX 6424: Data Mining to Advance Healthcare
Instructor Name
Month Day, Year
Introduction
Clinical questions often begin broadly: why do our patients with chronic obstructive pulmonary disease (COPD) keep coming back? Data can help answer such questions, but only after the question is narrowed to something the data can address. This paper takes that broad question and scopes it to the electronic health record of a 250-bed community hospital, showing each step of the narrowing, the variables available, the population the data can identify, the completeness of the data, and the limits of what can be concluded. No real patients or staff are identified.
The Narrowing Is Shown
The question was narrowed in four steps. Step 1, broad question: Why are patients with COPD readmitted? Step 2, measurable outcome: Which patients admitted for a COPD exacerbation are readmitted within 30 days? This defines the outcome as a 30-day readmission, a standard measure. Step 3, answerable with local data: Among patients discharged from our hospital after a COPD exacerbation, which characteristics documented during the stay are associated with readmission to our hospital within 30 days? This limits the outcome to readmissions the hospital can actually see. Step 4, final scoped question: Among adults aged 40 and older discharged alive from our hospital after an admission with a principal diagnosis of COPD exacerbation between January 2022 and December 2024, which characteristics recorded during the index admission are associated with unplanned readmission to our hospital within 30 days? Each step trades breadth for answerability, and the final question can be answered with data that exist, which the first question could not.
Variables at Field Level
Each variable was defined at the level of the record field that stores it.
| Variable | Record field | Notes |
|---|---|---|
| COPD exacerbation (index admission) | Principal diagnosis code, ICD-10 J44.1 | Excludes patients with COPD coded as secondary |
| Age | Date of birth, calculated at admission | Complete |
| Prior admissions in past year | Encounter table, inpatient class | Our hospital only |
| Home oxygen | Discharge medication and equipment orders | Missing if arranged by outside supplier |
| Length of stay | Admission and discharge timestamps | Complete |
| Discharge disposition | Discharge disposition field | Complete |
| Inhaler teaching documented | Nursing education flowsheet row | Inconsistently used |
| Follow-up appointment scheduled | Scheduling system appointment within 14 days | Our clinics only |
| Readmission within 30 days | Encounter table, unplanned inpatient admission | Our hospital only |
Population the Data Can Find
The data can find patients whose COPD exacerbation was coded as the principal diagnosis. They cannot find patients admitted for pneumonia whose COPD worsened, or patients admitted with a principal diagnosis of respiratory failure due to COPD, unless those codes are included. After reviewing coding practice with the health information management department, the definition was expanded to include respiratory failure codes when COPD was the first secondary diagnosis. The final population includes 1,284 index admissions over three years.
Completeness Reported
Completeness varies by variable. Age, length of stay, discharge disposition, and diagnosis codes are complete. Inhaler teaching is documented for only 41% of admissions, probably because nurses document teaching in free-text notes rather than the flowsheet row. Home oxygen is missing when arranged by outside suppliers, which the case management team estimates is about a third of cases. Follow-up appointments are captured only for the hospital's own clinics. Weiskopf and Weng (2013) describe completeness as one of several dimensions of data quality, along with correctness, concordance, plausibility, and currency, and they note that incomplete data can bias analyses in ways that are hard to detect.
Willing to Say No
Some parts of the original question cannot be answered with these data. Readmissions to other hospitals are invisible, so the analysis will undercount readmissions, perhaps substantially, since many patients live closer to other facilities. The data cannot show whether patients filled prescriptions, used inhalers correctly, or had home support, all of which likely affect readmission. It would be misleading to present the analysis as explaining readmission; it can only describe which recorded characteristics are associated with readmission to this hospital. A systematic review of readmission prediction models found that most performed poorly, partly because they lacked data on social and functional factors (Kansagara et al., 2011). This analysis shares that limitation.
Checking the Data Before Trusting Them
Before analysis, the data will be checked for plausibility and correctness as well as completeness. A random sample of 50 index admissions will be compared with the full chart to confirm the principal diagnosis, the discharge disposition, and any readmission. Length of stay values above 30 days will be reviewed individually, since they may reflect data entry errors or unusual cases that would distort averages. Readmissions coded as planned, such as scheduled procedures, will be confirmed and excluded. These checks take time, but an analysis built on unchecked data can reach confident conclusions that are simply wrong, and the checks themselves often reveal documentation problems worth fixing.
What the Analysis Can Contribute
Despite its limits, the analysis can contribute. It can identify characteristics, such as prior admissions or discharge to home without services, that are associated with readmission to this hospital. It can reveal documentation gaps, such as inhaler teaching, that the hospital can fix. And it can provide a baseline for evaluating a future transitional care program.
Nursing's Stake in the Question
The question matters to nursing because several of the variables most likely to be associated with readmission are nursing work: inhaler teaching, assessment of home support, and coordination of follow-up. If the analysis shows that patients with documented teaching and a scheduled follow-up are readmitted less often, the finding would support investment in nurse-led discharge education, while acknowledging that association does not prove the teaching caused the difference.
Next Steps
The next steps are to request the data extract through the hospital's data governance process, confirm that the analysis qualifies as quality improvement, and consult a statistician on the analytic approach. The documentation gap for inhaler teaching will be reported to the nursing practice council, with a suggestion that teaching be documented in the structured flowsheet row rather than in notes, so that future analyses can use it.
Conclusion
Scoping a broad question about COPD readmissions to the data available produced a precise, answerable question with defined variables, a known population, and reported completeness. The analysis will describe associations within this hospital's data and will say clearly what it cannot explain.
References
Kansagara, D., Englander, H., Salanitro, A., Kagen, D., Theobald, C., Freeman, M., & Kripalani, S. (2011). Risk prediction models for hospital readmission: A systematic review. JAMA, 306(15), 1688-1698. https://doi.org/10.1001/jama.2011.1515
Weiskopf, N. G., & Weng, C. (2013). Methods and dimensions of electronic health record data quality assessment: Enabling reuse for clinical research. Journal of the American Medical Informatics Association, 20(1), 144-151. https://doi.org/10.1136/amiajnl-2011-000681
NURS FPX 6424 Assessment 2 instructions, in plain terms
The instructions ask you to develop a clinical question that data mining could address and to assess whether available data can answer it. You state the question, identify the data needed, inventory what exists and where, report gaps and quality problems, revise the question to fit the data, define variables and the population, and assess feasibility. Showing the question changing is part of the task. Data sources are usually the electronic health record and related systems, described in general or composite terms. Scholarly sources on the clinical problem and on data quality support the paper. The output should be a question an analyst could start work on.
How the NURS-FPX6424 Assessment 2 example is put together
Initial question, requirements, inventory, gaps, revision, variables, population, feasibility. An initial question block states what the writer wanted to know. A requirements block says what data would be needed. An inventory block reports what exists, in which systems and over what period. A gaps block names what is missing or unusable. A revision block presents the question the data can support and explains what was given up. A variables block specifies each field, its values and its completeness. A population block defines who is included in terms the data can identify. A feasibility block states what the analysis will and will not be able to establish, before any analysis begins. Field completeness is reported as a percentage wherever it is known.
Where the marks sit in the NURS FPX 6424 Assessment 2 scoring guide
The scoring guide rates the clinical question, the data requirements, the data inventory, the gaps, the revised question and the variable and population definitions. The question reaches Distinguished when it is revised visibly in response to the data. Variables score higher when defined at field level with values and completeness. The population criterion rewards definitions the data can identify, such as diagnosis codes and encounter types. Gaps earn credit when their effect on the question is explained. Feasibility should be honest, including saying no. Writing and APA formatting are assessed as well. The annotations show where the question narrowed and what was given up at each step, which is the evidence of judgment the criterion is looking for.
NURS FPX 6424 Assessment 2 help: mistakes that cost marks
Students often write a question first and never check it against the data. Another common mistake is variables described loosely, such as smoking status, without saying where the field lives or how complete it is. Populations defined clinically may not be findable in the data. Data quality checks are frequently skipped. Some papers promise analyses the data cannot support. Check for problems like readmissions to other hospitals, which a single system's data cannot see, and say how that limits the question. Keep the organization's systems anonymous. A model data-scoped question on your own clinical problem can be drafted for you.
Get a NURS FPX 6424 Assessment 2 example written to your instructions
Send the Assessment 2 instructions and the scoring guide from your NURS FPX 6424 courseroom, plus the question and the data your own version scopes. We write a custom example against those exact criteria and return it in 24 to 48 hours. The first custom sample is free, and showing the narrowing from first question to answerable one is what this assessment is built to examine. The paper above is an original model document written by our desk, not a submitted student paper and not an official Capella University document.
More NURS FPX 6424 papers and MSN samples
- NURS FPX 6424 Assessment 1: Data Mining Vocabulary Grounded in Nursing Practice
- NURS FPX 6424 Assessment 3: Proposal to Administration
- NURS FPX 6424 Assessment 4: Practice Recommendation with Measure, Baseline and Target
- NURS FPX 6426 Assessment 2: Needs Assessment Meeting with Stakeholders
- NURS FPX 6422 Assessment 2: CITI Training
- NURS FPX 6620 Assessment 2: One Model Applied to a Named Population
- NURS FPX 6400 Assessment 2: Tracing One Clinical Workflow Into Its Data
NURS FPX 6424 Assessment 2 questions, answered
Why show the question changing?
Because scoping is what is being assessed. A question that appears already matched to the data suggests either luck or concealment, and the reasoning between the two versions is where the analytical work lives. Show what you wanted to know, what that would have required, what exists, and what you can therefore actually ask.
How specific do variables need to be?
Field level. Not mobility status but the named assessment item, its permitted values, where it lives and how often it is completed. Conceptual variables hide the problems: two fields may both appear to capture the same thing while being completed by different people at different times, and only the specific version surfaces that.
What if the data cannot answer my question?
Say so, and say what would be needed. That is a legitimate and useful finding, and it is far better than proceeding with a proxy that answers something adjacent while appearing to answer the original. Name what collection change or additional source would make the question tractable, which turns a dead end into a recommendation.
Where can I find a free NURS FPX 6424 Assessment 2 sample paper?
This page holds the full COPD readmission question scoped to EHR data, with field-level variables, population, completeness, data checks and next steps. A free custom paper on your clinical question is available.
What if my data cannot answer my NURS FPX 6424 Assessment 2 question?
Say so and revise the question to what the data can support, explaining what was given up. Showing that judgment is part of what the assessment rewards.