HA535 Unit 8 Utilization Metrics Analysis Example

Quickest Becomes Slowest Once Case Mix Counts: A Utilization Metrics Analysis of a Year of Bed Use at a Composite System's Three Hospitals, From Occupancy and Length of Stay to Observation Shares and Avoidable Days

[Student Name]

Purdue University Global

HA535: Data Analytics for Health Care Managers

Unit 8 Utilization Metrics Analysis

[Instructor Name]

[Date]

The system, the hospitals and every metric are composite and bracketed for the writer's own data.

Data Sources and Definitions

All figures cover one fiscal year of inpatient discharges drawn from the cleaned extract built in Unit 2, after the observation miscoding at the rural hospital was corrected. Staffed beds are the beds open and crewed on an average day, not licensed beds, because licensed beds overstate what a hospital can actually use. Average daily census is total inpatient days divided by 365. Observation stays are counted separately and never added to inpatient days, so occupancy here reflects inpatient use only. Avoidable days come from the case management log, where a case manager records the date a physician documents medical readiness and the date the patient leaves; every day between them counts as avoidable unless a clinical change is noted.

Occupancy

Occupancy = average daily census ÷ staffed beds. Flagship: [287.3] ÷ [326] = [88.1] percent. Suburban: [115.1] ÷ [150] = [76.6] percent. Rural: [21.6] ÷ [52] = [41.6] percent. The flagship runs close to the level at which waits for beds rise sharply; the rural hospital has room.

Average Length of Stay, Raw and Adjusted

Raw average length of stay: flagship [5.45] days, suburban [5.10], rural [4.53]. Case mix index, which scores how resource-heavy each hospital's mix of diagnoses is, with [1.00] as the national middle: [1.78], [1.31], [0.98]. Case-mix-adjusted length of stay = raw length of stay ÷ case mix index: flagship [3.06], suburban [3.89], rural [4.63]. The rural hospital treats less complex patients, so a short raw stay hides the fact that its patients stay longer than their conditions would predict. Adjusting for case mix is standard practice because hospitals treating sicker patients will naturally have longer stays; comparing raw figures rewards treating simpler cases (Cleverley & Cleverley, 2018).

Observation Stays

Share of all stays billed as observation: flagship [18.3] percent, suburban [21.8], rural [34.3]. The rural hospital's high share follows partly from the miscoding corrected in Unit 2 and partly from its role caring for patients who need short monitoring rather than admission. Because observation stays are excluded from inpatient length of stay, a high share can make inpatient stays look longer by removing short ones.

Avoidable Days

Avoidable days are days when patients are medically ready to leave but wait for a post-acute bed, such as skilled nursing or rehabilitation. System total: [7,120] days, [4.6] percent of patient days, about [19.5] beds held every day of the year. The flagship holds [4,960] of them. Research on delayed discharge links these waits to longer stays, functional decline and frustration for patients and staff (Rojas-García et al., 2018).

Putting the Metrics Together

The flagship runs near full occupancy, treats the most complex patients efficiently on a case-mix basis and holds most of the system's avoidable days. Freeing even half of its avoidable days would return about [6.8] beds a day, the equivalent of a small unit. The rural hospital has spare beds and longer adjusted stays, which suggests opportunities to accept transfers from the flagship for appropriate patients.

How Rankings Moved

By raw length of stay, the ranking was rural, suburban, flagship. By adjusted length of stay, it reverses: flagship, suburban, rural. Leaders who used raw figures would have praised the wrong hospital.

Questions Each Hospital Should Answer Next

The flagship should ask which post-acute providers account for most of its waits and whether a placement specialist would shorten them. Its leaders should also check whether patients ready on Fridays wait longer, since many skilled nursing facilities admit less over weekends. The suburban hospital sits in the middle on every measure; its main question is whether its observation share, the highest outside the rural site, reflects appropriate use or a habit of placing patients in observation while a decision is delayed. The rural hospital should review a sample of long stays with its physicians to find whether the cause is waiting for specialist input from the flagship, limited weekend discharge or something else. Each hospital should bring its answer to the next system operations meeting, so that the comparison turns into action rather than a ranking.

Recommendations

1. Report case-mix-adjusted length of stay, not raw, in system comparisons.

2. Track avoidable days by hospital and reason weekly, on a run chart so that real changes can be separated from week-to-week noise (Provost & Murray, 2011).

3. Explore transfers of suitable flagship patients to the rural hospital.

4. Investigate why rural adjusted stays are long, including physician coverage and discharge processes.

Limits

Case mix index captures diagnosis but not every aspect of severity. Avoidable days depend on how consistently case managers record medical readiness, which varies by hospital.

Conclusion

Rankings move once metrics are defined properly. Adjusted for case mix, the flagship is the most efficient, the rural hospital the least, and the system's real capacity problem is the [19.5] beds a day held by patients waiting to leave.

References

Cleverley, W. O., & Cleverley, J. O. (2018). Essentials of health care finance (8th ed.). Jones & Bartlett Learning.

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

Rojas-García, A., Turner, S., Pizzo, E., Hudson, E., Thomas, J., & Raine, R. (2018). Impact and experiences of delayed discharge: A mixed-studies systematic review. Health Expectations, 21(1), 41-56. https://doi.org/10.1111/hex.12619