Lupus & antiphospholipid syndrome
Development and temporal validation of a clinical prediction model to identify bacterial infection in hospitalised patients with systemic lupus erythematosus
Authors' conclusion
A routinely available, interpretable model identified BI among hospitalised patients with SLE using NA SLE as the comparator.
Study context
- Population
- Hospitalised patients with SLE admitted for bacterial infection (BI) or active disease, January 2023 to January 2026395 training, 200 validation admissions
- Comparison
- Bacterial infection vs non-infected active SLE
- Outcome
- Identification of BI (model discrimination, AUC)
Methodological detailsLASSO regression across imputed datasets, multivariable logistic regression, full-workflow bootstrap internal validation, same-centre temporal validation, decision curve analysis
- LASSO regression across imputed datasets
- Multivariable logistic regression
- Full-workflow bootstrap internal validation
- Same-centre temporal validation
- Decision curve analysis
Performance
As reported in the abstract
Model discrimination (AUC), training cohort
Training AUC of the model
0.869
95% CI 0.833 to 0.905
AUC of C reactive protein (CRP)
0.771
AUC of procalcitonin
0.682
AUC of ESR/CRP ratio
0.736
Further findings
Full-workflow bootstrap internal validation
- Optimism-corrected AUC
- 0.837
- Corrected Brier score
- 0.162
Same-centre temporal validation
- AUC
- 0.852 (95% CI 0.794 to 0.910)
- Acceptable calibration and preserved net benefit
Original publication
Development and temporal validation of a clinical prediction model to identify bacterial infection in hospitalised patients with systemic lupus erythematosus. Lupus Sci Med. Published 3 September 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →