Admission-day SOFA-2 versus Sequential Organ Failure Assessment (SOFA) for mortality discrimination: external validation in a large multicenter intensive care unit (ICU) database
Authors' conclusion
SOFA-2 showed similar discrimination for ICU mortality compared with SOFA-1 and maintained broadly comparable performance across clinically relevant subgroups
Study context
- Population
- Adult ICU stays, OneICU Japanese critical care database, February 2013 to August 2025
- Participants
- 152,883 eligible ICU stays
- Exposure
- SOFA-2 vs SOFA-1 on ICU admission day
- Outcome
- Discrimination for ICU mortality (AUROC)
Methodological detailsexternal validation, multicentre database
- External validation
- Multicentre database
Main finding
As reported in the abstract
AUROC for ICU mortality
| Group | Value | AUC |
|---|---|---|
| SOFA-2 | 0.859 | 0.859 |
| SOFA-1 | 0.853 | 0.853 |
Value
SOFA-20.859
SOFA-10.853
AUC
SOFA-20.859
SOFA-10.853
- SOFA-2 vs SOFA-1: P < 0.001 — DeLong test; absolute difference small
Observational data: associations, not effects.
Further findings
Discrimination, mechanical circulatory support subgroups (subgroup)
- SOFA-2 showed higher discrimination than SOFA-1
Discrimination, other device-defined subgroups and ICU readmissions (subgroup)
- Discrimination was similar
SOFA-2 discrimination across major diagnostic groups (subgroup)
- SOFA-2 demonstrated good discrimination
- 121,443 stays with complete SOFA-2 data
Original publication
Admission-day SOFA-2 versus SOFA for mortality discrimination: external validation in a large multicenter ICU database. Crit Care. Published 12 April 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →