Systematic overestimation of mortality by Simplified Acute Physiology Score (SAPS) II in contemporary Intensive Care Unit (ICU) practice: a retrospective study analyzing ICU patient outcome using the 2013-2024 French National Health Database
Authors' conclusion
Even if SAPS II maintains adequate discrimination, its systematic overprediction undermines its utility for contemporary critical care practice
Study context
- Population
- Adult ICU admissions, French National Health Database, 2013–2024
- Participants
- 2,702,204 ICU admissions
- Exposure
- Original SAPS II vs Quintic model
- Outcome
- Calibration (Brier score) of in-hospital mortality prediction
Methodological detailsnationwide hospital billing database, derivation and validation sets, 80% (n = 2,161,763) derivation set; 20% (n = 540,441) validation set; mean SAPS II = 41 ± 21.3
- Nationwide hospital billing database
- Derivation and validation sets
- 80% (n = 2,161,763) derivation set; 20% (n = 540,441) validation set; mean SAPS II = 41 ± 21.3
Main finding
As reported in the abstract
Predicted vs observed in-hospital mortality (validation set)
- Observed mortality rate in validation set
- 20.0%
- Relative improvement in overall Brier score with new calibration
- more than 19%
- Relative improvement in Brier score for intermediate SAPS II values (60-79)
- more than 29%
Observational data: associations, not effects.
Further findings
| Outcome | Original SAPS II | Quintic model |
|---|---|---|
| Discrimination (AUC) | ||
| AUC | 0.830 | 0.832 |
Discrimination (AUC)
- Discrimination remains similar
Original publication
Systematic overestimation of mortality by SAPS II in contemporary ICU practice: a retrospective study analyzing ICU patient outcome using the 2013-2024 French National Health Database. Anaesth Crit Care Pain Med. Published 24 September 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →