HINF 500 Week 3 How Health Data Are Collected, Used and Reported Example

The Life of One Timestamp: Following a Sepsis Antibiotic Time From the Bedside to a Federal Website, and Why the Hospital's Score Was Lower Than Its Care

[Student Name]

University of Phoenix

HINF/500: Informatics for Health Administration

Week 3 Assignment

[Instructor Name]

[Date]

The hospital, its scores and the patient are composites written for a model paper; national data and research findings come from the sources listed.

One line was printed in red on the quarterly quality report that reached the hospital's operations chief: the hospital's score on the federal measure for early management of severe sepsis and septic shock was 58%. Nationally, hospitals averaged 65% for October 2024 through September 2025, and the top tenth reached 83% (Centers for Medicare & Medicaid Services, 2026). The chief nursing officer insisted that the emergency department gave antibiotics quickly, and the emergency physicians agreed. To find out, the vice president asked to follow a single data point through the system: the time a sepsis patient received antibiotics.

The Measure

The federal sepsis measure is a bundle: within set time windows after sepsis is recognized, the hospital must draw a lactate level, obtain blood cultures before antibiotics, give broad-spectrum antibiotics, give fluids for low blood pressure or high lactate, repeat the lactate if the first was elevated and document reassessment in septic shock. A case passes only if every element is met and documented. The measure is reported by hospitals to the federal quality reporting program and published on a public comparison website.

Step One: Collection at the Bedside

A 71-year-old man arrived at the emergency department with fever, confusion and low blood pressure. At 10:42 p.m., a nurse scanned the barcode on his antibiotic bag and on his wristband, and the electronic record stored the administration time automatically. Barcode scanning made this timestamp more reliable than a handwritten note. Other elements, such as the time sepsis was recognized, depend on a clinician's documentation, and are less reliable.

Step Two: Storage

The record stores the administration as a structured field in its clinical database, copied nightly to the hospital's data warehouse, where analysts combine it with admission, laboratory and billing data. Structured fields, such as the scan time, can be queried; free-text notes, where much of the sepsis story lives, cannot without manual review or text processing.

Step Three: Abstraction

The federal measure is not calculated automatically. A trained quality abstractor reviews a sample of eligible cases each month, following detailed specifications. For the man's case, she identified time zero, the time sepsis criteria were met and documented, as 9:58 p.m. His antibiotics at 10:42 p.m. were well within three hours. But the case failed: his first lactate was elevated, and the repeat lactate was drawn 6 hours and 40 minutes after time zero, beyond the 6-hour window. The documentation also lacked a clear time for the repeat draw, which the abstractor had to infer from the laboratory's collection record.

Step Four: Internal Use

Abstracted results feed the hospital's quality dashboard, reviewed monthly by the sepsis committee, and a unit-level report showing which bundle elements fail most often. Before this review, the dashboard showed only the overall pass rate, not the reasons.

Step Five: External Reporting

The hospital submits abstracted data quarterly to the federal reporting program. After validation, results are published on the public comparison website, where patients, insurers and journalists can see them, and they can affect payment and the hospital's reputation.

What the Review Found

Reviewing all 64 failed cases from the past two quarters, the quality team found that antibiotic timing caused only 9 failures. The most common causes were late or missing repeat lactates (27 cases), incomplete documentation of fluid volume or reassessment (18 cases) and disputed time zero because of vague documentation (10 cases). The hospital was not slow to treat sepsis; it was slow to finish and document the bundle.

Why the Data Source Matters

The same patients can be counted in very different ways. In a study of 409 hospitals, sepsis incidence measured with clinical criteria from electronic records was stable from 2009 to 2014, while incidence measured with claims codes rose about 10% a year, and mortality trends also differed by source (Rhee et al., 2017). Claims reflect changes in coding practices as well as disease. Administrators comparing sepsis figures across years or hospitals must ask where the numbers come from.

Why Timing Still Matters

The measure's emphasis on speed has a clinical basis. New York made sepsis protocols mandatory in 2013, and researchers then examined almost 50,000 patients treated under that mandate. Every added hour before the three-hour bundle was finished, and every added hour before antibiotics, carried roughly 4% higher odds of dying in the hospital after risk adjustment, though slower fluid boluses showed no such link (Seymour et al., 2017). Fast antibiotics protect patients; complete documentation protects the score that reports it. For the hospital, the finding also meant that its strength, quick antibiotics, was the element with the strongest link to survival, a point worth making to staff discouraged by the red number on the report.

Fixes to Documentation, Abstraction and Reporting

The hospital adopted four changes. The electronic record now fires an automatic order for a repeat lactate when the first is elevated, with a timed reminder to the nurse. A structured sepsis note replaces free text for time zero, fluid volume and reassessment. Abstractors meet monthly with emergency physicians to review disputed cases. And the dashboard now reports failure reasons by bundle element, updated weekly from a preliminary automated screen.

Who Owns Each Step

Following the timestamp revealed that no single person owned the data flow. Nurses owned bedside documentation, the information technology department owned the record and warehouse, the quality department owned abstraction and submission and the emergency department owned the care. Each group could see its own step but not the others. The vice president assigned the sepsis committee's physician chair as the owner of the measure from end to end, with a data steward in the quality department responsible for definitions, abstraction consistency and the weekly report.

Automated Measurement on the Horizon

Federal programs are moving from manually abstracted measures toward electronic clinical quality measures calculated directly from structured record data. Automation would reduce abstraction costs and allow every case, not a sample, to be measured. But it depends on exactly the kind of structured, timely documentation this review found missing. The hospital's new structured sepsis note is therefore also preparation for a future in which the record, not an abstractor, produces the score.

Limits

The changes could raise the score without changing outcomes if documentation improves faster than care. The sepsis committee will therefore also track mortality for patients with sepsis identified by clinical criteria, not only the bundle score.

Conclusion

One antibiotic timestamp, scanned at 10:42 p.m., passed through collection, storage, abstraction, internal review and public reporting. It was accurate, and the patient was treated quickly, yet the case failed because other elements were late or poorly documented. Understanding how data are produced let the hospital fix the right problems, and research on data sources and timing showed why both measurement and care matter.

References

Centers for Medicare & Medicaid Services. (2026). Timely and effective care: National [Data set]. Provider Data Catalog. https://data.cms.gov/provider-data/dataset/isrn-hqyy

Rhee, C., Dantes, R., Epstein, L., Murphy, D. J., Seymour, C. W., Iwashyna, T. J., Kadri, S. S., Angus, D. C., Danner, R. L., Fiore, A. E., Jernigan, J. A., Martin, G. S., Septimus, E., Warren, D. K., Karcz, A., Chan, C., Menchaca, J. T., Wang, R., Gruber, S., & Klompas, M. (2017). Incidence and trends of sepsis in US hospitals using clinical vs claims data, 2009-2014. JAMA, 318(13), 1241-1249. https://doi.org/10.1001/jama.2017.13836

Seymour, C. W., Gesten, F., Prescott, H. C., Friedrich, M. E., Iwashyna, T. J., Phillips, G. S., Lemeshow, S., Osborn, T., Terry, K. M., & Levy, M. M. (2017). Time to treatment and mortality during mandated emergency care for sepsis. New England Journal of Medicine, 376(23), 2235-2244. https://doi.org/10.1056/NEJMoa1703058