Document Type : Original Article
Authors
1
Department of Health Policy and Management, Seoul National University College of Medicine, Seoul, Republic of Korea
2
Institute of Health Policy and Management, Seoul National University Medical Research Center, Seoul, Republic of Korea
3
Department of Education and Human Resource Development, Seoul National University Hospital, Seoul, Korea
4
Department of Preventive Medicine, Dankook University Hospital, Cheonan, Republic of Korea
5
Clinical Preventive Medicine Center, Seoul National University Bundang Hospital, Seongnam, Republic of Korea
Abstract
Background
Unnecessary hospitalizations undermine the efficient use of healthcare resources and may expose patients to preventable safety risks. This study aimed to develop a computerized screening tool for identifying potentially inappropriate adult inpatient admissions using nationwide health insurance claims data in South Korea.
Methods
We conducted a retrospective cross-sectional study using the Health Insurance Review and Assessment Service-National Inpatient Sample (HIRA-NIS) (2017–2019) to develop and assess a claims-based hospitalization appropriateness screening tool (HAST) for adult admissions. Based on the adult hospitalization clinical service criteria from the Appropriateness Evaluation Protocol (AEP), we selected only those criteria directly observable in claims data and operationalized them using standardized electronic data interchange codes. The final tool included nine criteria and episodes meeting none of these criteria were classified as potentially inappropriate admissions (PIAs). HAST was assessed through a single-round of expert review for content validity, mortality-based known-groups comparison, and
convergent validity against selected SQLape diagnostic groups. Logistic regression models were applied to identify factors associated with PIA.
Results
The final tool comprised nine screening criteria: (1) receiving general anesthesia, (2) admission to intensive care unit (ICU), (3) receiving hospice care, (4) childbirth, (5) continuous monitoring of vital signs, (6) receiving arterial blood gas analysis (ABGA), (7) application of mechanical ventilator, (8) receiving intravenous injections, and (9) receiving intramuscular injections. Across analytic specifications, 17.5% to 19.2% of admissions were classified as PIAs. PIAs were more frequent among adults aged 20–39 years, females, National Health Insurance (NHI) beneficiaries, and patients admitted to small facilities with fewer than 100 beds.
Conclusion
HAST offers a feasible, reproducible, and scalable method for monitoring the appropriateness of hospital admissions using routinely collected administrative data. It can be used for large-scale surveillance to support health policy evaluation and system-level quality improvement.
Keywords