NR 534 Week 7 Performance Scorecard Review Example

Performance Scorecard Review: Reading a Nine-Site Sepsis Scorecard by Asking Where Each Patient Was When the Outcome Was Decided, and Why the Critical Access Hospitals Look Best on the Measure They Hand to Someone Else

Student Name

College of Nursing, Chamberlain University

NR 534: Healthcare Systems Management

Instructor Name

Month Day, Year

Performance Scorecard Review: Reading a Nine-Site Sepsis Scorecard by Asking Where Each Patient Was When the Outcome Was Decided, and Why the Critical Access Hospitals Look Best on the Measure They Hand to Someone Else

The Scorecard

Our nine-site health system reports a quarterly performance scorecard to its board. The sepsis section shows four measures by site: in-hospital mortality for patients with sepsis, completion of the sepsis bundle within three hours, intensive care length of stay and 30-day readmission. On last quarter's scorecard, the two critical access hospitals had the lowest sepsis mortality in the system, 4.2% and 5.0%, and the tertiary center had the highest, 16.0%. A board member asked whether the tertiary center should learn from the critical access hospitals.

This review answers that question by reading every shared measure with one question in mind: where was the patient when the outcome was decided, and where was it counted? When a critical access hospital sends its sickest patient to the tertiary center, that patient's outcome is counted at the tertiary center, not at the hospital where the patient arrived, was recognized and was first treated.

Where the Patients Went

Table 1 follows the sepsis patients who first presented at each type of site last quarter.

Table 1

Sepsis Patients by Site of First Presentation, Last Quarter

What the Table Shows

The critical access hospitals kept 44 of their 58 sepsis patients and transferred 14. Those they kept were mostly less sick; those they transferred were the ones on vasopressors or needing ventilation. Of the 14 transferred, 5 died at the tertiary center. Counted at the site of first presentation, the critical access hospitals' mortality is 12.1% (7 of 58), not 4.5%. The freestanding departments transfer almost everyone, so their own-site mortality is meaningless as a measure of their care.

The tertiary center's reported mortality of 16.0% combines its own direct presentations with 93 patients transferred in from other sites, who were among the sickest in the system. For patients who presented directly to the tertiary center, mortality was 15.2%. The board member's comparison was therefore between hospitals that send away their sickest patients and the hospital that receives them.

This problem is well described. A simulation study found that transferring even a small number of additional critically ill patients out of a hospital could substantially lower its standardized mortality ratio, biasing inferences about quality (Kahn et al., 2007). Our scorecard counts outcomes where they are recorded, so it rewards sites for transferring and penalizes the site that receives.

The Other Measures

Bundle completion. Bundle completion within three hours is measured at the site where the patient presented, which is appropriate. The critical access hospitals' completion was 71%, below the community hospitals' 84%, largely because antibiotics were delayed while blood cultures were drawn by a single nurse. This measure is correctly attributed and points to a real opportunity at the smaller sites, where a second person to draw cultures would let antibiotics start sooner.

Intensive care length of stay. The tertiary center's longer stays partly reflect the patients it receives. The measure should be adjusted for severity or reported separately for transferred patients.

Readmission. Thirty-day readmission is counted at the discharging hospital. For transferred patients, that is the tertiary center, although many are discharged home to the communities of the sending hospitals, whose follow-up care determines much of the readmission risk.

Transfer timing. The scorecard has no measure of how long transfers take. A national review concluded that organizational structures to route critically ill patients to the right hospital reliably have not developed, and that the resulting delays may cost lives (Iwashyna, 2012). Our critical access hospitals' median time from transfer decision to departure is 2 hours and 50 minutes. That interval belongs to no site on the scorecard.

Recommendations

First, report sepsis mortality by site of first presentation, counting every patient who presented at a site regardless of where they died. This measures the whole system's care of patients who came through each door and removes the incentive to look better by transferring.

Second, add a transfer measure for sending sites: the minutes between agreeing to transfer and the patient leaving, reported by site and owned jointly by the sending site and the system transfer center.

Third, keep bundle completion at the site of presentation, where it belongs, and set a specific target for the critical access hospitals with support from the tele-critical care program.

Fourth, report intensive care length of stay and readmission separately for transferred and direct patients at the tertiary center, so that the tertiary center is not judged against sites that send it their sickest patients.

Fifth, add a short note on the scorecard explaining attribution, so that board members read each figure with the right question.

Why This Matters for a System

A single hospital can be judged on its own outcomes. A system's hospitals share patients, and a scorecard that ignores that sharing misleads the board about where care is good and where it needs help. Research on hospital mergers has found that joining a system does not by itself improve quality (Beaulieu et al., 2020). If our scorecard is to show whether our system is making care better, it has to follow patients across buildings. The critical access hospitals are not failing; their bundle completion needs work, and their decision to transfer the sickest patients is appropriate. But they should not be presented as the system's best performers on a measure they hand to someone else, and the tertiary center should not be asked to learn from a comparison that was never fair. The board member's question was a good one, and it deserves a scorecard that can answer it. With mortality counted by site of first presentation, the board would see that the community hospitals, not the critical access hospitals, currently have the lowest mortality for all patients who come through their doors, and could ask what they do differently. That is a comparison worth learning from.

References

Beaulieu, N. D., Dafny, L. S., Landon, B. E., Dalton, J. B., Kuye, I., & McWilliams, J. M. (2020). Changes in quality of care after hospital mergers and acquisitions. New England Journal of Medicine, 382(1), 51-59. https://doi.org/10.1056/NEJMsa1901383

Iwashyna, T. J. (2012). The incomplete infrastructure for interhospital patient transfer. Critical Care Medicine, 40(8), 2470-2478. https://doi.org/10.1097/CCM.0b013e318254516f

Kahn, J. M., Kramer, A. A., & Rubenfeld, G. D. (2007). Transferring critically ill patients out of hospital improves the standardized mortality ratio: A simulation study. Chest, 131(1), 68-75. https://doi.org/10.1378/chest.06-0741