NU806 Unit 2 DNP Results Section Example

From 14.3 to 9.0 Percent, With an Interval That Crosses Zero: A Unit 2 Results Section for an Insulin Discharge Readiness Project

[Student Name]

Purdue University Global

NU806: DNP Project III

Unit 2 Results Section

[Instructor Name]

[Date]

The hospital, units, patients and results are composite and illustrative; the methods sources cited are real.

Patients

Results are reported in the order SQUIRE 2.0 suggests, from the people and process to the outcomes and any unintended effects (Ogrinc et al., 2016). From go-live on each unit to the data cutoff, 214 patients were discharged home on new insulin from the eight units. The main analysis covers 144 patients from the earliest four units, the two medical-surgical units that opened the spread and the cardiac and surgical units that followed, since only those units had six or more months of data. Their baseline comparison group is the 223 patients discharged on new insulin from the same four units in the 12 months before each unit's start. The 70 patients from the last four units, oncology, renal-transplant, orthopedic and observation, are reported descriptively. No eligible patient was excluded from analysis, and no patient declined the bundle entirely, though eleven declined the class.

The main-analysis patients had a median age of 63 years, and 41 of the 144, 28 percent, had Medicaid or no insurance, compared with 59 of 223, 26 percent, at baseline. Thirty-nine lived more than 30 miles from the hospital.

Delivery of the Bundle

Component by component, 136 of the 144 gave a watched dose with their own pen (94 percent), 142 had a written plan (99 percent), 127 held the pen by their first insulin day (88 percent) and 134 were reached by day three (93 percent). A class or virtual class seat was accepted by 58, 40 percent, and 39 attended. All five parts were delivered to 117 patients, 81 percent, against the target of 85 percent.

Monthly run charts by unit, each annotated with go-live dates and the eleven logged changes as recommended for improvement data (Provost & Murray, 2011), show the pattern over time. On 4 West and 3 West, the all-five share climbed from 60 percent in the first month and then held between 82 and 91 percent from month five to month nine. The pen measure's run chart for those two units showed seven months in a row above the pre-bundle median, enough to count as a shift (Perla et al., 2011). No shift or trend was detected on the outcome run chart.

Delivery met the run chart's test for a real change; the outcome did not.

Outcome

Thirteen of the 144, 9.0 percent, came back within 30 days through the emergency department or as an admission with a coded glucose diagnosis: hypoglycemia, hyperglycemia or ketoacidosis. At baseline, 32 of 223 returned, 14.3 percent. The absolute difference was 5.3 points; its 95 percent interval stretched from a drop of 11.9 points to a rise of 1.2. The chi-square statistic was 2.3 on one degree of freedom (p about 0.13).

Of the 13 returns, seven were for hyperglycemia, three for hypoglycemia and three for diabetic ketoacidosis. Five occurred in the first seven days after discharge, and four of those five involved patients who had replied to the day-one text without reporting a problem.

Balancing Measures

Patients left a median of 2 hours 36 minutes after the discharge order was signed, four minutes sooner than in the baseline year. Nurse time on the follow-up queue averaged 2.1 hours a week across all eight units once all were live, below the three-hour review threshold. Returns for hypoglycemia were 3 of 144, 2.1 percent, against 6 of 223, 2.7 percent, at baseline. The stopping rule, three or more hypoglycemia returns in one month, was never triggered.

Stratified Results

A rate appears below only where a group reached 20 patients; smaller groups are given as counts. Among the 41 Medicaid or uninsured patients, all five parts were delivered to 30, 73 percent, and 6 returned, 14.6 percent. Among the other 103 patients, all five parts were delivered to 87, 84 percent, and 7 returned, 6.8 percent. Among the 39 patients living more than 30 miles away, 11 attended a class or virtual class, 28 percent, against 28 of 105 living closer, 27 percent. Fourteen patients were 80 or older; 13 received contact by day three and two returned.

The Correction Scale Finding

Among patients discharged with a correction scale who replied on day one, 8 of 13 answered "not sure how much" before the teaching change in week 12, and 5 of 31 after it.

The Last Four Units

Of the 70 patients from oncology, renal-transplant, orthopedic and observation, followed for between two and five months, 58 received all five parts, 83 percent, and 7 returned within 30 days, 10.0 percent. Their baseline, from the year before, was 25 of 180, 13.9 percent. These figures are descriptive only.

Cost

Actual project costs were about $74,900, against a budget of $78,400, mostly because resource nurse audit time fell after the first months.

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

Provost, L. P., & Murray, S. K. (2011). The health care data guide: Learning from data for improvement. Jossey-Bass.