3-STEP MODEL- AN EXPLORATIVE NOVEL APPROACH TO CLASSIFY SEPSIS: A LONGITUDINAL OBSERVATIONAL STUDY

Introduction: Sepsis remains a critical healthcare challenge worldwide, demanding prompt identification and treatment to improve patient outcomes. Given the absence of a definitive gold standard diagnostic test, there is an imperative need for adjunct diagnostic tools to aid in early sepsis detection and guide effective treatment strategies. This study introduces a novel 3-step model to identify and classify sepsis, integrating current knowledge and clinical guidelines to enhance diagnostic precision. Methods: This longitudinal observational study was conducted at a tertiary care teaching hospital in northern India. Adult patients admitted with suspected sepsis underwent screening using predefined criteria. The 3-step model consisted of Step 1, assessing dysregulated host response using a National Early Warning Score-2 (NEWS-2) score of ≥6; Step 2, evaluating risk factors for infection; and Step 3, confirming infection presence through clinical, supportive, or confirmatory evidence. Patients were categorized into Asepsis, Possible sepsis, Probable sepsis, or Confirmed sepsis at various intervals during hospitalization. Results: A total of 230 patients were included. Initial categorization on Day 1 showed 13.0% in Asepsis, 35.2% in Possible sepsis, 51.3% in Probable sepsis, and 0.4% in confirmed sepsis. By Day 7, shifts were observed with 49.7% in Asepsis, 9.5% in Possible sepsis, 25.4% in Probable sepsis, and 15.4% in confirmed sepsis. At discharge or death, categories were 60.4% Asepsis, 5.2% Possible sepsis, 21.7% Probable sepsis, and 12.6% Confirmed sepsis. Transitions between categories were noted throughout hospitalisation, demonstrating the dynamic nature of sepsis progression and response to treatment. Conclusion: The 3-step model effectively stratifies sepsis status over hospitalization, facilitating early identification and classification of septic patients. This approach holds promise for enhancing diagnostic accuracy, guiding clinical decision-making, and optimizing antibiotic stewardship practices. Further validation across diverse patient cohorts and healthcare settings is essential to confirm its utility and generalizability.