A simple electronic medical record-based predictors of illness severity in sepsis (sepsis) score

• DOI: 10.1371/journal.pone.0299473
• Publication Date: 2024-06-26T17:44:20Z
• Authors: Alex M. Cressman; Bijun Wen; Sudipta Saha; Hae Young Jun; Riley Waters; Sharan Lail; Aneela Jabeen; Radha Koppula; Lauren Lapointe-Shaw; Kathleen A. Sheehan; Adina Weinerman; Nick Daneman; Amol A. Verma; Fahad Razak; Derek MacFadden
• Keywords: Medical record

Objective Current scores for predicting sepsis outcomes are limited by generalizability, complexity, and electronic medical record (EMR) integration. Here, we validate a simple EMR-based score in large multi-centre cohort. Design A record-based predictor of illness severity (SEPSIS) was developed (4 additive lab-based predictors) using population-based retrospective cohort study. Setting Internal medicine services across four academic teaching hospitals Toronto, Canada from April 2010—March 2015 (primary cohort) 2015–2019 (secondary cohort). Patients We identified patients admitted with based upon receipt antibiotics positive cultures. Measurements main results The primary outcome in-hospital mortality secondary were ICU admission at 72 hours, hospital length stay (LOS). calculated the area under receiver operating curve (AUROC) SEPSIS score, qSOFA, NEWS2. then evaluated (2015–2019) hospitalized receiving antibiotics. Our included 1,890 median age years (IQR: 56–83). 9% died during hospitalization, 18.6% to ICU, mean LOS 12.7 days (SD: 21.5). In (2015–2019, 4811 patients) cohorts, AUROCs 0.63 0.64 respectively, which similar NEWS2 (0.62 0.67) qSOFA 0.68). > 14 days, between scores, cohorts. All had comparable calibration mortality. Conclusions An shows ability predict important clinical compared other validated (qSOFA NEWS2). Because score’s simplicity, it may prove useful tool research applications.