HI230 Unit 8 Seminar Reflection on Length of Stay Example

Who Left the Count: A Unit 8 Seminar Reflection on an Average Length of Stay That Rose From 4.3 to 4.8 Days Without Any Patient Staying Longer, and the Question I Now Ask Before Reading Any Rising Rate

[Student Name]

Purdue University Global

HI230: Quality Assurance and Statistics in Health Information 1

Unit 8 Seminar Reflection

[Instructor Name]

[Date]

The hospital, its service and its figures are composite; the seminar exchange is reconstructed, and the sources cited are real.

What Happened in the Seminar

Our instructor put two lines on the screen for a composite hospital's adult medical service. In the first year, 9,800 discharges averaged 4.3 days. In the second, 8,300 discharges averaged 4.8 days. The question was simple: what changed? I answered first, and quickly. Half a day longer per patient looked to me like discharges being held up, perhaps by a shortage of skilled nursing beds or a slower pharmacy at the end of the stay. I had read about discharge delays in another course and the explanation fit the number. I said it with some confidence.

The First Explanation and Why It Felt Right

Looking back, my answer was not foolish. A longer average stay often does mean slower discharges, and I had seen exactly that pattern on a medical floor during a clinical rotation, where patients waited days for a nursing home bed. The trouble was that I read the change in the average as a change in care before asking whether the group being averaged had stayed the same. I noticed the numerator had moved and never looked at the denominator, even though it had shrunk by 1,500 patients right in front of me. I read a change in the average as a change in care. The first thing that had changed was the list of patients being averaged.

Who Left the Count

A classmate asked the question that undid my answer: who were the 1,500 discharges that disappeared? The instructor then showed us that the hospital's largest payer had changed its policy that year, and roughly 1,500 one- and two-day stays had moved from inpatient admission to observation status. Observation patients are outpatients in billing and in hospital statistics, so they vanish from inpatient discharges and inpatient days alike, even when they occupy the same bed on the same floor (Horton, 2017). Observation stays had been rising for years among Medicare patients, which made the shift more plausible than my story about discharges (Feng et al., 2012), and a study at one academic center found observation stays common and often lasting more than one night, which is exactly the kind of stay that used to count as a short admission (Sheehy et al., 2013). The recalculation settled it. The first year carried 42,140 inpatient days. Taking out 1,500 short stays averaging a day and a half removes about 2,250 days, leaving 39,890 days over 8,300 discharges, an average of 4.8. The second year's 8,300 discharges carried 39,840 days. The two years were nearly identical once they described the same kind of patient. Nobody stayed longer. The shortest stays had left the calculation, and the average rose because of who was left.

So What: Other Ways a Count Can Change Underneath a Rate

The same reclassification moved other figures that year. The service's gross death rate rose from 2.0 to 2.4 percent with almost the same number of deaths, because the shorter stays that left the denominator were the patients least likely to die. A committee shown that figure alone might have started a mortality review for a problem that existed only in the arithmetic. I can now see other ways membership shifts. A new hospice unit can raise a hospital's death rate by bringing in patients admitted to die. A surgical service that moves its simple cases to an outpatient center will see its inpatient average rise and its complication rate climb, because the remaining patients are sicker. A coding rule that changes which diagnoses count as complications can move an infection rate with no change at the bedside. An aging catchment area slowly raises almost every rate. In each case the rate is computed correctly and still misleads anyone who assumes the group behind it stayed the same.

Now What: The Question I Will Ask

My change in practice is specific. Before I hand any committee a rate that has moved, I will put the denominator beside it for both periods and ask whether its membership changed: a new unit, a new payer rule, a new coding guideline, a transfer of patients to another setting. If the denominator moved by more than a few percent, I will recalculate the earlier period on the same basis before anyone discusses causes. I expect this will sometimes show a real change in care, and that is fine. The point is to know which kind of change I am reporting. What I learned about myself is less comfortable. I answered first because I wanted to be the one with the explanation, and the classmate who asked a question did more useful work than I did with my answer. In my next seminar I intend to ask what changed in the count before I offer a reason for the change, which is a reflective habit Rolfe and colleagues describe as moving from describing an event to deciding what to do differently because of it (Rolfe et al., 2001).

References

Feng, Z., Wright, B., & Mor, V. (2012). Sharp rise in Medicare enrollees being held in hospitals for observation raises concerns about causes and consequences. Health Affairs, 31(6), 1251-1259. https://doi.org/10.1377/hlthaff.2012.0129

Horton, L. A. (2017). Calculating and reporting healthcare statistics (5th ed.). AHIMA Press.

Rolfe, G., Freshwater, D., & Jasper, M. (2001). Critical reflection for nursing and the helping professions: A user's guide. Palgrave Macmillan.

Sheehy, A. M., Graf, B., Gangireddy, S., Hoffman, R., Ehlenbach, M., Heidke, C., Fields, S., Liegel, B., & Jacobs, E. A. (2013). Hospitalized but not admitted: Characteristics of patients with "observation status" at an academic medical center. JAMA Internal Medicine, 173(21), 1991-1998. https://doi.org/10.1001/jamainternmed.2013.8185