NU 476 Unit 6 outcome measurement example

Outcome Measurement: Nurse-Initiated Standing Orders for Inpatient Influenza Vaccination on a Medical Unit

Student Name

Department of Nursing, Herzing University

NU 476: Nursing Capstone

Instructor Name

March 25, 2026

Outcome Measurement: Nurse-Initiated Standing Orders for Inpatient Influenza Vaccination on a Medical Unit

The Measure and the Two Periods

The project introduced a nurse-initiated standing order protocol for influenza vaccination on 4 North, a 30-bed adult medical unit: nurses screened every patient on admission for vaccination status and contraindications and, for eligible unvaccinated patients who consented, gave the vaccine before discharge without a separate prescriber order. The approach is supported by evidence that standing orders increase inpatient vaccination compared with reminders to physicians (Dexter et al., 2004), and influenza vaccination is recommended for everyone 6 months and older without contraindications (Grohskopf et al., 2024).

Primary measure: the proportion of eligible patients discharged from 4 North who were vaccinated before discharge. Eligible means adult, not documented as vaccinated this season, no documented contraindication, and a stay of at least 24 hours. Baseline period: October 13 to November 7 of the current season, four weeks before the protocol. Implementation period: November 10 to December 5, four weeks after. The same method was used for both periods: I reviewed every discharge from the unit in each period using the same data fields, and a second reviewer checked 20 records from each period with 100% agreement.

Results

Baseline: 156 discharges, 88 eligible, 42 vaccinated before discharge, 48%. Implementation: 149 discharges, 83 eligible, 59 vaccinated, 71%. The absolute increase was 23 percentage points. Process measures: admission screening for vaccination status was documented for 94% of patients in the implementation period compared with 61% at baseline. Of eligible patients not vaccinated during implementation, 14 declined after discussion, 6 were missed at discharge, and 4 were discharged before the vaccine could be given. The number declining was similar in both periods, about one in six eligible patients.

Could the Difference Be Chance?

With 88 and 83 eligible patients, some variation between periods would occur even if nothing had changed. A simple comparison of two proportions gives a z statistic of about 3.1 for the difference between 48% and 71%, giving a two-tailed p of roughly 0.002, so chance alone is an unlikely explanation. That calculation assumes the two periods are independent samples from the same kind of patients, which is only approximately true, and it says nothing about whether the project caused the difference. It answers a narrower question: whether the rise is larger than the usual week-to-week noise on a unit of this size. The more important questions, whether something other than the project produced the change, are taken up next.

Is the Change Caused by the Project?

Three alternative explanations were considered. First, seasonal timing: the baseline period fell earlier in the influenza season, when more patients might already be planning to be vaccinated elsewhere, and the implementation period fell closer to the holidays and rising influenza activity, when patients might be more willing. This could explain part of the increase. A useful check is the hospital's two other medical units, which did not change practice; their rates rose from 45% to 51% over the same periods, suggesting that seasonal and general factors account for a few percentage points, not most of the change. Second, attention: nurses knew the project was being measured, which may have increased effort during the four weeks. Whether the change holds after the project ends is a separate question, addressed in the handover plan. Third, patient mix: the proportion of eligible patients was similar in both periods, 56% and 56%, and the median age was similar.

What the Results Mean

The most reasonable reading is that the protocol increased inpatient vaccination on 4 North by about 15 to 20 percentage points beyond what seasonal factors produced, with the process data showing how: screening became routine, and vaccination no longer depended on a prescriber remembering to order it. The result falls short of the project's target of 80%, and the data show why: missed doses at discharge and early discharges. Those are the next steps, for example adding vaccination status to the discharge checklist, to be tested in a further short cycle before the protocol is spread (Langley et al., 2009). Reporting the shortfall with its reasons is more useful to the unit than reporting only the improvement.

Limits

Four weeks per period on one unit is a small sample, and the comparison units are not a randomized control. Vaccination given elsewhere after discharge was not captured. The result applies to this unit and this season; whether it would hold on units with different patient populations would need its own measurement.

References

Dexter, P. R., Perkins, S. M., Maharry, K. S., Jones, K., & McDonald, C. J. (2004). Inpatient computer-based standing orders vs physician reminders to increase influenza and pneumococcal vaccination rates: A randomized trial. JAMA, 292(19), 2366-2371. https://doi.org/10.1001/jama.292.19.2366

Grohskopf, L. A., Ferdinands, J. M., Blanton, L. H., Broder, K. R., & Loehr, J. (2024). Prevention and control of seasonal influenza with vaccines: Recommendations of the Advisory Committee on Immunization Practices, United States, 2024-25 influenza season. MMWR. Recommendations and Reports, 73(5), 1-25. https://doi.org/10.15585/mmwr.rr7305a1

Langley, G. J., Moen, R. D., Nolan, K. M., Nolan, T. W., Norman, C. L., & Provost, L. P. (2009). The improvement guide: A practical approach to enhancing organizational performance (2nd ed.). Jossey-Bass.