Sepsis and septic shock case identification from electronic health records: an open-source workflow and comparison of cohorts by criteria
Authors' conclusion
Agreement between case criteria was fair to moderate.
Study context
- Population
- Patient encounters 2012–2024 across 25 EDs and hospitals in an integrated health system; first sepsis encounter302,112 sepsis
- Exposure
- CDC ASE vs Sepsis-3 vs International Classification of Diseases
- Outcome
- Pairwise agreement between criteria (Cohen's kappa and Jaccard index)
Methodological detailsopen-source workflow, multicentre, integrated health system, Programming code for the complete workflow available on a public platform
- Open-source workflow
- Multicentre
- Integrated health system
- Programming code for the complete workflow available on a public platform
Main finding
As reported in the abstract
Pairwise agreement between criteria
κ
0.39
sepsis; Jaccard index = 0.31; highest pairwise agreement
κ
0.53
septic shock; Jaccard index = 0.37; highest pairwise agreement
- Multiple criteria satisfied, patients meeting ≥1 sepsis criterion
- 35%
- Multiple criteria satisfied, patients meeting ≥1 septic shock criterion
- 45%
Observational data: associations, not effects.
Further findings
Patient characteristics and outcomes by criteria
- Characteristics and outcomes differed across cohorts identified with the three criteria
- Patients meeting both CDC ASE and Sepsis-3 criteria more severely ill, with high in-hospital mortality incidence
Original publication
Sepsis and septic shock case identification from electronic health records: an open-source workflow and comparison of cohorts by criteria. Crit Care. Published 12 December 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →