| Course | DNP 960A DPI Project: Part II |
|---|---|
| Topic | Topic 4 |
| Paper type | Data collection report |
| 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 960A Topic 4
Expected Against Received: A Three-Month Data Collection Report for Statin Outreach, Including Missing Scores That Cluster in Evening Calls
[Author Name]
College of Nursing and Health Care Professions, Grand Canyon University
DNP-960A: DPI Project: Part II
Topic 4 Assignment
[Instructor Name]
December 10, 2029
The organization, setting and figures below are a composite written as a model document. No real employer, client, colleague or patient is described.
Expected Against Received: A Three-Month Data Collection Report for Statin Outreach, Including Missing Scores That Cluster in Evening Calls
What This Report Covers
This report covers data collected for the statin outreach project from September through November 2029. It reports what arrived, not what was planned, and it sets expected and received counts side by side so that no shortfall is averaged away. The extracts arrived on the scheduled day each month, so the run chart has no gaps (Perla et al., 2011). The report covers four things: the monthly registry extracts that feed the primary and new-starts measures, the call log and fidelity checklist, the belief questionnaire scores and the decisions taken about records that did not fit the plan. Every decision about data described here was made under a rule written in the measurement plan before implementation, except one, which is identified as such.
Expected and Received, by Month
Table 1
Expected and Received Data, September Through November 2029
| Source | September expected / received | October expected / received | November expected / received |
|---|---|---|---|
| Registry extract rows (cohort) | 971 / 971 | 971 / 968 | 971 / 917 |
| Completed first calls | 55 / 48 | 55 / 61 | 55 / 52 |
| Fidelity checklists | 48 / 48 | 61 / 59 | 52 / 47 |
| First-call questionnaires | 48 / 43 | 61 / 50 | 52 / 38 |
| Follow-up calls due / completed | 0 / 0 | 40 / 29 | 55 / 36 |
| Follow-up questionnaires | 0 / 0 | 29 / 24 | 36 / 25 |
Table 1 compares expected and received data for each source and month.
The follow-up rows show the fidelity problem predicted in the monitoring plan. Of 95 follow-up calls due in October and November, 65 were completed, 68 percent, just under the 70 percent threshold that triggers a meeting; the meeting took place on December 3, and the daily task list for follow-ups was switched on the next week.
Missingness, by Field and by Group
The first-call questionnaire is the field with the most missing values: 131 of 161 completed first calls have both scale scores, 81 percent, falling from 90 percent in September to 73 percent in November. Broken down by group, the pattern is not random. Of the 30 calls without scores, 21 were made after 5 p.m., when care managers call patients who work during the day. Evening calls make up about a third of first calls but account for 70 percent of missing questionnaires. The care managers explained that evening patients are often calling from a car or a break room and agree to discuss the decision but not to answer ten additional questions.
This matters for analysis. A value missing completely at random does not bias results; a value missing for a reason related to the patient does (Sterne et al., 2009). Evening patients are younger, more often employed full-time and, in the call log, more likely to decline a statin: 38 percent of evening patients declined, against 22 percent of daytime patients. If their questionnaires are missing, the mechanism analysis will overrepresent daytime patients, who are more likely to start. The measurement plan's rule stands: only patients with a questionnaire at both calls enter the mechanism analysis, and below a 70 percent completion threshold that analysis becomes descriptive. At the current rate that threshold will be missed, and the report will say that the patients without scores differ systematically from those with them.
Two responses are being tried in December without changing the rule. The care managers will offer evening patients a shortened version of the call in which the questionnaire is sent by text message for completion afterward, and the lead care manager will track whether scores returned this way differ from scores collected by phone. Neither response will be used to fill in the missing September to November scores, which stay missing.
Duplicates and Exclusions
Duplicates were identified with a rule written into the measurement plan: a patient may be counted as a new start only once in any 90-day window, using the first order or fill. The rule found 14 duplicates in October, mostly patients whose pended order was signed and then reissued at a different dose after the pharmacy flagged an interaction, and 6 in November. All were removed from the new-starts count and the rule is stated in the monthly report.
Two records were excluded from the cohort under a rule also written in advance: patients who develop a contraindication after the cohort date leave the denominator from the month it is recorded. One patient became pregnant in October; one had a myocardial infarction in November and moved into secondary prevention, which is outside the project's population. Both exclusions are counted and reported, and neither was discretionary.
A Reorganization That Changed the Population
In November the network transferred Clinic E's adult patients to a new federally qualified health center that opened nearby, under an agreement between the two organizations. Patients who moved lost their attribution to the network, and 50 cohort patients left the registry in the November extract, which explains most of the drop from 971 to 917 rows. The measurement plan anticipated patients leaving attribution individually, but not a whole clinic's worth at once, so this is the one decision not covered by a prewritten rule. After discussion, the network's quality director and I agreed on November 20 that the 50 patients will be removed from both the baseline and implementation periods of the paired comparison, so the comparison covers the same patients throughout, and that Clinic E will be reported separately for September and October. Combined attrition, counting the transfer, the two exclusions and individual departures, is now 6 percent, below the 10 percent at which the plan requires a repeated analysis. Because a whole clinic left for reasons unrelated to the project, the comparison will still be reported both ways, with and without Clinic E's pre-transfer months, so readers can see that the choice does not drive the result (Ogrinc et al., 2016).
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
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, Article b2393. https://doi.org/10.1136/bmj.b2393
Reading the GCU DNP 960A Topic 4 instructions
Topic 4 in DNP-960A asks students to report on data collection during implementation. Strong reports compare what was expected with what was received for each source and month, so gaps are visible, and break missingness down by field and by group, since an overall figure can hide missing data concentrated in one group. Students should apply duplicate and exclusion rules written in advance and report how many cases each rule affected. Any change to the population during the project, such as a reorganization, should be reported with its effect on the denominator and the plan for handling it, and the report should state what the data can still support.
Inside the DNP 960A Topic 4 example
The finished example reports arrivals rather than intentions. Expected and received counts sit side by side for each period, and the shortfall is not averaged away. Missingness is broken out by field and by group, because a value absent at random is a different problem from one absent on nights, and the second pattern turns out to hold here. Duplicates are identified with the rule that identified them. A service reorganization changes the eligible population partway through, and the example states the decision taken about the affected period and who agreed it. Two records were excluded under a rule written before the data was seen, and the report says so. One planned comparison is flagged as now at risk, with what would rescue it.
Where the marks sit in the DNP 960A Topic 4 rubric
Faculty reward expected and received set side by side. An Excellent report tables each source by month and shows follow-up calls falling behind as predicted. Credit goes to breaking missing questionnaire scores down by group, finding 131 of 161 complete and gaps clustered in evening calls, with Sterne and colleagues on multiple imputation risks. The rubric values a 90-day duplicate rule that removed 14 cases. Strong reports explain Clinic E's patients moving to a new health center and adjust the denominator. Top marks go to SQUIRE-ready reporting of each change.
DNP 960A Topic 4 help: mistakes that cost marks
The failure this course punishes hardest is reporting the collection as designed rather than as the site has permitted it to run, and here that means a report restating the protocol with no arrival figures in it. A second weakness is a single completeness percentage, which conceals whether the absence sits in one group and therefore whether the comparison survives. Marks also go for deleting incomplete records silently, since an exclusion rule invented after the data is seen cannot be defended to anybody. A denominator change left unreported invalidates every rate around it. Extracts with no named owner cannot be repeated. Reports with nothing at risk claim a smoothness that data collection at a live site rarely delivers.
Get a DNP-960A Topic 4 example written to your instructions
Send the DNP-960A Topic 4 assignment instructions and the rubric your classroom posts, with any reporting template your section supplies. We write a custom example to those criteria, with arrival set against expectation, missingness broken out by group, exclusion rules dated and one analysis flagged as at risk, in 24 to 48 hours. The first one is free.
More DNP 960A papers and DNP samples
- DNP 960A Topic 1: Site Readiness Report
- DNP 960A Topic 2: Early Implementation Report
- DNP 960A Topic 3: Fidelity Monitoring Plan
- DNP 960A Topic 5: Protocol Deviation DQ Pair
- DNP 960A Topic 6: Stakeholder Engagement Analysis
- DNP 960A Topic 7: Interim Data Review
- DNP 960A Topic 8: Plan Revision Report
- DNP 965A Topic 4: Limitations Statement
- DNP 955 Topic 4: Literature Arranged by Argument
- DNP 966 Topic 4: Site Continuity Report
- DNP 840 Topic 4: Influence Without Formal Power
DNP 960A Topic 4 questions, answered
Why break missingness down by group?
Because an overall figure answers the wrong question. Ten percent missing spread evenly costs precision; ten percent missing entirely from one shift or one service can remove the very comparison the project was built to make. The breakdown takes minutes and changes what the report can honestly claim, which is why reviewers look for it before they look at the completeness total.
Do you collect or supply any of my data?
Never. The data belongs to your site and your project, and it is yours to collect, hold and analyze under the permissions you obtained. A sample can only show the shape of the written report around it: how arrival is set against expectation, how missingness is broken out, and how a denominator change is disclosed rather than absorbed into a rate.
What if my denominator changes partway through?
Report the change, the date and the decision taken about the affected period, then say who agreed it. Reorganizations, closures and service mergers happen and are not fatal. What is fatal is a rate that silently spans two different eligible populations, because every comparison built on it becomes indefensible the moment a reader learns what happened.
Where can I find a free DNP 960A Topic 4 sample paper?
The data collection report for DNP 960A Topic 4 covers what it covers, expected and received by month, missingness by field and group, duplicates and exclusions and a reorganization that changed the population. Your own DNP 960A Topic 4 prompt gets a free report.
Why break missingness down by group in DNP 960A Topic 4?
Because an overall rate hides where data are missing. Questionnaire gaps clustered in evening calls, which reach a different group of working patients.