HLTH5633 Module 5: sample paper, in real form

Reducing Falls With Injury on a 36-Bed Adult Medical Department: A Measure Set, Three Improvement Cycles, and the Governance to Hold the Gain

[Author Name]

Master of Health Administration Program, American College of Education

HLTH5633 Advanced Quality Management for Healthcare Administrators

Module 5 Assignment

[Faculty Name]

August 11, 2026

Model document written for teaching. The hospital, the department and every figure are a composite; no real organization, employee or patient is described.

The Department, the Measure Set and the Baseline

Riverbend Regional is a composite 240-bed community hospital, and this analysis covers one department inside it: a 36-bed adult medical floor that recorded 11,400 patient days in the 12 months ending March 2026. Over that period the department reported 71 patient falls, a rate of 6.2 per 1,000 patient days, of which 18 caused injury, a rate of 1.6 per 1,000 patient days. Comparison medical departments in the same national database sat near 3.5 and 0.9 on those two measures. The department was not slightly behind its peers. It was running at close to twice the injury rate of the comparison group, on a measure that carries payment consequences under the federal hospital-acquired condition program (Centers for Medicare and Medicaid Services, 2024).

One number cannot direct improvement work, so the department adopted a set of four. The outcome measure is falls with injury per 1,000 patient days, chosen over a raw count because census swings by season, and chosen over total falls because injury is what the patient actually carries home. Two process measures track the changes being tested: a documented fall risk screening completed within four hours of admission and repeated at every shift change, and documented overnight rounding. The fourth is a balancing measure, described later, because any department can lower its fall rate by keeping patients in bed. Each measure carries a written definition, a numerator, a denominator and a named owner, so the figures stay comparable as staff and seasons turn over.

Stratifying the baseline mattered more than the average did. Thirty eight of the 71 falls involved unassisted toileting, 61 percent happened between 7 p.m. and 7 a.m., and 27 involved patients taking a diuretic, a sedative or both. Only 54 percent of charts carried a fall risk screening repeated at shift change, though 96 percent carried the admission screening, which pointed at a follow-through problem rather than a knowledge problem. That pattern fits a hospital falls literature that credits multifactorial programs over any single device or alarm (LeLaurin & Shorr, 2019). Post-fall reviews existed but were completed a day or more after the event, by which time the detail worth having was gone.

Applying the Model for Improvement

The department used the Model for Improvement, which asks what the team is trying to accomplish, how a change will be recognized as an improvement, and what changes will produce it, then runs each change through plan-do-study-act cycles (Langley et al., 2009). The aim was written with a number and a date: reduce falls with injury on the department from 1.6 to 0.8 per 1,000 patient days by June 30, 2027, with no loss in documented patient mobility. That method was chosen over a full define-measure-analyze-improve-control project because the department did not need a new root cause study. It needed small, fast tests of changes it could already name, and the Model for Improvement is built for testing at that scale.

Three cycles ran in sequence. The first tested scheduled toileting rounds overnight with six high-risk patients across ten days, staffed by two nurses and one aide; documentation held at 91 percent, no tested patient fell, and the change moved forward. The second cycle spread the round to the whole overnight shift for 30 days and added a bedside card showing each patient's mobility level, which surfaced a supply problem when three rooms turned out to have no working bed exit alarm. The third cycle moved the post-fall huddle to within 60 minutes of the event, with a structured five-question review, so the learning reached the next shift rather than the next monthly meeting.

Results were plotted on a run chart by month rather than judged by comparing two averages, because one good quarter proves very little in a department this size (Provost & Murray, 2011). Eight consecutive points below the baseline median satisfied the shift rule, the signal that something other than chance has moved the process. Falls with injury fell from 1.6 to 0.7 per 1,000 patient days over the following eight months, while total falls moved less, from 6.2 to 4.4. That difference is worth naming honestly: the department got better at protecting patients who fall before it got better at preventing falls. The measure set makes that visible; a single headline number would have concealed it.

Balancing Measure, Governance and Sustainment

The balancing measure was documented ambulation episodes per patient day, watched alongside restraint hours per 1,000 patient days as a second check. Falls programs fail quietly in exactly this way: a department leans on alarms, keeps patients in bed, records fewer falls, and discharges weaker patients after longer stays. Ambulation held at 2.1 episodes per patient day against a 2.0 baseline and restraint hours did not rise, which is the finding that makes the outcome believable. Had ambulation dropped, the result would have been reported as a trade in harms rather than a gain, and the department would have kept the toileting round while giving up the alarm work that arrived with it.

Governance is what separates a project from a practice. The department quality council reviews all four measures monthly with the department director as owner, and results roll into the nursing quality committee and then to the board quality committee each quarter, where falls with injury appears on a dashboard beside its peer benchmark rather than alone. An escalation trigger is written down in advance: two consecutive months above 1.0 per 1,000 patient days returns the measure to active project status with a named leader and a 60-day plan. The Joint Commission (2015) treats a fall with serious injury as a reviewable sentinel event, which is why this measure reaches the board at all rather than staying inside nursing operations.

Sustainment rests on standard work rather than on attention. The toileting round and the 60-minute post-fall huddle were written into department standard work, added to orientation for new staff, and audited on ten records a month by the charge nurse, with results posted where staff can see them. Two limits belong in any honest report. A single department across eight months cannot rule out seasonal staffing effects, and self-reported process measures overstate performance in every organization that uses them. The department accepts both, which is why the outcome measure and the balancing measure, not the process measures, decide whether the change is kept (Agency for Healthcare Research and Quality, 2024).

References

Agency for Healthcare Research and Quality. (2024). Preventing falls in hospitals: A toolkit for improving quality of care. U.S. Department of Health and Human Services. https://www.ahrq.gov/patient-safety/settings/hospital/fall-prevention/toolkit/index.html

Centers for Medicare & Medicaid Services. (2024). Hospital-acquired conditions. U.S. Department of Health and Human Services. https://www.cms.gov/medicare/quality

The Joint Commission. (2015). Sentinel event alert 55: Preventing falls and fall-related injuries in health care facilities. https://www.jointcommission.org/resources/sentinel-event/sentinel-event-alert-newsletters/

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.

LeLaurin, J. H., & Shorr, R. I. (2019). Preventing falls in hospitalized patients: State of the science. Clinics in Geriatric Medicine, 35(2), 273-283.

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