External validation of Sequential Organ Failure Assessment (SOFA) 2 score: a retrospective Chinese cohort study
Authors' conclusion
In this external validation, SOFA-2 demonstrated improved discrimination for ICU mortality and more consistent score distribution than SOFA-1.
Study context
- Population
- Adults admitted to intensive care units (ICUs) in China, 2014–2024; SOFA calculated from first 7 ICU days
- Participants
- 118,542 adults
- Exposure
- SOFA-2 vs SOFA-1 score
- Outcome
- ICU mortality (discrimination by AUROC)
Methodological detailsmulticentre, external validation, Multiple approaches applied to address missing data
- Multicentre
- External validation
- Multiple approaches applied to address missing data
Main finding
As reported in the abstract
Discrimination for ICU mortality
| Group | Value | AUROC |
|---|---|---|
| SOFA-2 | 80.9% (95% CI 80.3%–81.5%) | 80.9% |
| SOFA-1 | 76.7% (95% CI 76.0%–77.4%) | 76.7% |
Value
SOFA-280.9% (95% CI 80.3%–81.5%)
SOFA-176.7% (95% CI 76.0%–77.4%)
AUROC
SOFA-280.9%
SOFA-176.7%
Observational data: associations, not effects.
Further findings
| Outcome | OR (95% CI) | |
|---|---|---|
| ICU mortality per 1-point increase in SOFA-2 | 1.33 (1.33–1.35) | |
| 0.512Odds ratio (log scale) |
| Outcome | SOFA-2 | SOFA-1 |
|---|---|---|
| Between-cohort varianceIndicating greater robustness across these cohorts | 0.07 | 0.23 |
ICU and hospital mortality (overall cohort)
- Died in ICU
- 5,286 (4.45%)
- Died during hospitalisation
- 6,173 (5.2%)
SOFA-2 trajectories over first 7 days
- Widening divergence between survivors and non-survivors
SOFA-2 subscores and total scores
- Higher SOFA-2 subscores (0-4) and total scores associated with lower frequency and higher ICU mortality
Parts of this Brief are withheld: the automatic checks could not confirm them against the abstract.
Original publication
External validation of SOFA 2 score: a retrospective Chinese cohort study. Ann Intensive Care. Published 25 June 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →