HA425 Unit 10 ED Final Improvement Report Example

From Sixty-Four Minutes to Forty-One: A Final Improvement Report on a Composite Community Hospital Emergency Department's Evening Flow, Assembling the Analysis, the Two Changes and Twelve Weeks of Evaluation With Balancing Measures and Remaining Problems

[Student Name]

Purdue University Global

HA425: Operational Analysis and Quality Improvement

Unit 10 Final Improvement Report

[Instructor Name]

[Date]

The department, the weekly results and every figure are composite and bracketed; the report is a course model that draws together the earlier units.

Executive Summary

Problem: evening arrivals waited much longer than daytime arrivals at the same demand, and [6.4] percent left without being seen. Cause: the nursing template closed four rooms at 19:00, during the arrival peak. Changes: one shift moved to 11:00 to 23:00; a six-recliner results-pending area opened. Result: evening median door-to-provider time fell from [64] to [41] minutes; evening walkouts fell to [2.7] percent; the morning median rose from [18] to [22] minutes, within the limit set in advance.

Part 1: Analysis in Brief

Process map (Appendix A): of the nine handoffs mapped, the room assignment and the moment results reached the provider accounted for most of the waiting. Data analysis (Appendix B): October's door-to-provider times split by arrival hour into two operations, with the evening median more than double the day's. Capacity analysis (Appendix C): Little's Law showed evening demand for about [22] occupied rooms against [20] staffed after 19:00. Root cause analysis (Appendix D): an ordinary evening's walkout traced to a staffing template never compared with the arrival curve.

Part 2: Intervention

The shift change was introduced in three plan-do-study-act cycles: a volunteer on Tuesdays and Wednesdays for two weeks, all weekdays from week 3, and weekends from week 9. Four nurses volunteered for the rotation; no one was assigned. The results-pending area opened in week 6, one week later than planned, because the tracking-board location change slipped. It was tested with up to three patients a night for two weeks and opened fully in week 8. Both changes worked on the same constraint from different sides: one added staffed rooms, the other shortened the time each patient held one.

Part 3: Evaluation

Run chart of weekly evening medians (Figure 1): twelve baseline weeks centered on [64] minutes. The first week after the change sat near the baseline median at [61], because the volunteer test covered only two evenings. The next eleven weeks all fell below the baseline median, ranging from [38] to [52], a run that signals a real shift by standard run chart rules (Perla et al., 2011). The final four weeks averaged [41] minutes. Evening walkout rate (Figure 2): from [6.4] percent at baseline to [2.7] percent in the last four weeks. Process measures: staffed rooms at 20:00 rose from [20] to [24] on every weekday after week 3. Patients moved to recliners spent a median of [38] fewer minutes in treatment rooms. Balancing measure (Figure 3): the morning median door-to-provider time rose from [18] to [22] minutes, below the [10]-minute rise set as the pause trigger. No safety events were linked to the results-pending area.

Was the Aim Met?

The aim was a median under [45] minutes and walkouts under [3] percent within twelve weeks. Both were met by week 12. The evening median remains about ten minutes longer than the day's, so the evening is improved but not yet equal.

Comparison With Prediction

The capacity model predicted that [24] rooms and shorter room times would bring evening room use near [90] percent, similar to the day. Observed waits fell by about the amount the model suggested. Matching staffing to arrivals produced the kind of improvement queueing studies have reported for emergency departments (Green et al., 2006).

What Staff Reported

At the week 12 meeting, evening nurses said the 20:00 hour felt different: rooms were available, and the waiting room no longer filled with patients triaged hours earlier. Providers said the recliners worked best when a nurse moved patients promptly after the evaluation was complete; on two nights the area sat empty while rooms were full because no one had made the move. Day nurses reported the morning felt busier but manageable. The unit clerk asked for a printed list of recliner criteria at her desk, which was added in week 10.

Costs

Recurring: about [$13,140] a year in evening differential. One time: about [$18,000] for the recliner area. Walkouts fell by about [3.7] percentage points of evening arrivals, or roughly [190] patients a year who are now seen.

Remaining Problems

1. Provider utilization in the evening rose to about [90] percent; providers are now the next constraint.

2. Weekend evenings improved less, partly because weekend staffing has its own template.

3. Boarding of admitted patients still holds rooms on some nights, a hospital-wide issue that crowding research identifies as a major output factor (Hoot & Aronsky, 2008).

4. Results are still noticed late on some nights; a results alert to providers is under discussion.

Sustainability

Sustaining the gain depends less on enthusiasm than on structure. The 11:00 to 23:00 shift is now part of the permanent schedule template. The nurse manager owns a monthly review of evening medians and walkouts, and the quality committee receives a quarterly report. If the evening median rises above [50] minutes for four weeks, the team will reconvene.

Lessons Learned

The most visible queue, at triage, was not the constraint. Data split by hour revealed a problem that monthly averages hid. Testing with volunteers before changing the schedule built support, and a written prediction for each cycle made results easy to read (Langley et al., 2009).

What Did Not Go as Planned

The results-pending area opened a week late. The first volunteer cycle covered only two evenings a week, so its effect was hard to see. Weekend results lagged because the weekend schedule is built separately. None of these changed the outcome, but each lengthened the time before the effect was visible, which future projects should plan for.

Next Steps

Study provider coverage from 17:00 to 23:00, using the same capacity method that found the room constraint; review weekend templates with the staffing office; and pilot a results alert to providers on two evenings a week before deciding whether to spread it.

Conclusion

Two modest changes aimed at a constraint identified through careful analysis reduced evening waits by about a third and halved walkouts, at a small cost to the morning. The work also showed where the next constraint lies, which is where the department's next improvement cycle should begin. The same sequence of map, data, capacity model and root cause can be repeated for that next constraint.

References

Green, L. V., Soares, J., Giglio, J. F., & Green, R. A. (2006). Using queueing theory to increase the effectiveness of emergency department provider staffing. Academic Emergency Medicine, 13(1), 61-68. https://doi.org/10.1197/j.aem.2005.07.034

Hoot, N. R., & Aronsky, D. (2008). Systematic review of emergency department crowding: Causes, effects, and solutions. Annals of Emergency Medicine, 52(2), 126-136. https://doi.org/10.1016/j.annemergmed.2008.03.014

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.

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