Abstract: FR-PO0401
Administrative Severity Discordance in AKI
Session Information
- AKI: Biomarkers, Diagnostics, and Risk Prediction
October 23, 2026 | Location: Exhibit Hall A, Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Acute Kidney Injury
- 102 AKI: Clinical, Outcomes, and Trials
Authors
- Tai, Brent, BayCare Health System, Clearwater, Florida, United States
- Mittal, Ajay, BayCare Health System, Clearwater, Florida, United States
- D'Cunha, Prakas Thomas, BayCare Health System, Clearwater, Florida, United States
Background
Administrative severity classifications are widely used for inpatient risk adjustment and benchmarking. We evaluated whether discordance between administrative severity and clinical intervention intensity identifies heterogeneous inpatient risk patterns among hospitalized adults.
Methods
Retrospective study using 2022 National Inpatient Sample. Administrative severity defined using APR-DRG Severity of Illness: low (1–2) vs high (3–4). Clinical intervention intensity defined by dialysis and/or mechanical ventilation. Hospitalizations categorized as concordant low, discordant, or concordant high. Outcomes: mortality and LOS. Sensitivity analyses included AKI restriction, exclusion of ESRD/chronic dialysis, and alternative severity definitions.
Results
Adjusted mortality increased across phenotypes from 0.2% in concordant low to 2.6% in discordant and 16.0% in concordant high hospitalizations. Adjusted LOS similarly increased from 3.6 to 10.7 days. Findings were consistent in sensitivity analyses. Among AKI admissions, mortality increased from 0.4% to 29.9% across phenotypes.
Conclusion
Administrative severity and clinical intervention intensity represent distinct illness dimensions. Integration of clinical care escalation signals with administrative severity measures may improve future inpatient risk stratification and health system benchmarking.
Acknowledgment
We wanted to acknowledge all the HCUP Data Partners that contribute to HCUP. A link to the HCUP-US web page that contains the list of State organizations is here. (hcup-us.ahrq.gov/db/hcupdatapartners.jsp).
Mortality and LOS by discordance phenotype
| Phenotype | Adjusted mortality, % (95% CI) | Adjusted LOS, days (95% CI) |
| Concordant low | 0.2% (0.1%–0.3%) | 3.6 (3.3–4.0) |
| Discordant | 2.6% (1.7%–3.9%) | 6.5 (5.3–7.7) |
| Concordant high | 16.0% (13.7%–18.5%) | 10.7 (9.4–12.0) |
Mortality by discordance phenotype
Clinical intervention intensity by administrative severity category