A risk-adjusted length of stay to evaluate severe Community-Acquired Pneumonia (sCAP) outcomes: A machine learning analysis of 16,985 ICU admissions
Authors' conclusion
SLOSR shows promise as tool for assessing adjusted LOS as a surrogate of resource use in sCAP patients in the context of Brazilian ICUs.
Study context
- Population
- Adult sCAP admissions across 220 ICUs in 57 Brazilian hospitals (January-December 2023)
- Participants
- 16,985 ICU admissions
- Outcome
- Risk-adjusted LOS measured by machine learning-based Standardized Length of Stay Ratio (SLOSR = sum of observed LOS / sum of predicted LOS)
Methodological detailsmulticentre, machine learning model
- Multicentre
- Machine learning model
Main finding
As reported in the abstract
SLOSR model performance
Grouped R2
0.89
RMSE
4.57
MAE
3.10
Median Standardized Length of Stay Ratio (SLOSR) (funnel plot analysis; Q1 = 0.9; Q3 = 1.34)
1.13
Observational data: associations, not effects.
Further findings
Cohort characteristics
- Hospital mortality
- 9.3 %
- ICU mortality
- 6.4 %
- Median ICU length of stay (LOS) (days)
- 4
- Mean SAPS 3 score
- 50
- Received ventilatory support
- 28.1 %
Original publication
A risk-adjusted length of stay to evaluate severe Community-Acquired Pneumonia (sCAP) outcomes: A machine learning analysis of 16,985 ICU admissions. J Crit Care. Published 31 July 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →