Comparison of SIRS, qSOFA, and NEWS for the early identification of sepsis in the Emergency Department.

OBJECTIVES: The increasing use of sepsis screening in the Emergency Department (ED) and the Sepsis-3 recommendation to use the quick Sepsis-related Organ Failure Assessment (qSOFA) necessitates validation. We compared Systemic Inflammatory Response Syndrome (SIRS), qSOFA, and the National Early Warning Score (NEWS) for the identification of severe sepsis and septic shock (SS/SS) during ED triage. METHODS: This was a retrospective analysis from an urban, tertiary-care academic center that included 130,595 adult visits to the ED, excluding dispositions lacking adequate clinical evaluation (n = 14,861, 11.4%). The SS/SS group (n = 930) was selected using discharge diagnoses and chart review. We measured sensitivity, specificity, and area under the receiver-operating characteristic (AUROC) for the detection of sepsis endpoints. RESULTS: NEWS was most accurate for triage detection of SS/SS (AUROC = 0.91, 0.88, 0.81), septic shock (AUROC = 0.93, 0.88, 0.84), and sepsis-related mortality (AUROC = 0.95, 0.89, 0.87) for NEWS, SIRS, and qSOFA, respectively (p < 0.01 for NEWS versus SIRS and qSOFA). For the detection of SS/SS (95% CI), sensitivities were 84.2% (81.5-86.5%), 86.1% (83.6-88.2%), and 28.5% (25.6-31.7%) and specificities were 85.0% (84.8-85.3%), 79.1% (78.9-79.3%), and 98.9% (98.8-99.0%) for NEWS ≥ 4, SIRS ≥ 2, and qSOFA ≥ 2, respectively. CONCLUSIONS: NEWS was the most accurate scoring system for the detection of all sepsis endpoints. Furthermore, NEWS was more specific with similar sensitivity relative to SIRS, improves with disease severity, and is immediately available as it does not require laboratories. However, scoring NEWS is more involved and may be better suited for automated computation. QSOFA had the lowest sensitivity and is a poor tool for ED sepsis screening.

Selecting Test Cases from the Electronic Health Record for Software Testing of Knowledge-Based Clinical Decision Support Systems.

Software testing of knowledge-based clinical decision support systems is challenging, labor intensive, and expensive; yet, testing is necessary since clinical applications have heightened consequences. Thoughtful test case selection improves testing coverage while minimizing testing burden. ATHENA-CDS is a knowledge-based system that provides guideline-based recommendations for chronic medical conditions. Using the ATHENA-CDS diabetes knowledgebase, we demonstrate a generalizable approach for selecting test cases using rules/ filters to create a set of paths that mimics the system's logic. Test cases are allocated to paths using a proportion heuristic. Using data from the electronic health record, we found 1,086 cases with glycemic control above target goals. We created a total of 48 filters and 50 unique system paths, which were used to allocate 200 test cases. We show that our method generates a comprehensive set of test cases that provides adequate coverage for the testing of a knowledge-based CDS.