A study for comparison of severity assessment scoring systems in patients admitted in medical intensive care unit in a tertiary care hospital

Background: Accurate assessment of illness severity in critically ill patients is essential for effective clinical decision-making in intensive care units (ICUs). Various severity scoring systems, such as Acute Physiology and Chronic Health Evaluation (APACHE) II, APACHE IV, Simplified Acute Physiology Score II and III, Mortality Probability Model III (MPM III), Sequential Organ Failure Assessment (SOFA), Multiple Organ Dysfunction Score (MODS), and Logistic Organ Dysfunction Score (LODS), have been developed to predict outcomes. However, their comparative performance in resource-limited settings such as Indian tertiary care hospitals remains under-evaluated. Aims and Objectives: This study aims to compare the predictive accuracy of multiple ICU severity scoring systems in terms of mortality and morbidity among patients admitted to the medical ICU (MICU). Materials and Methods: This prospective observational study enrolled 250 adult patients admitted to the MICU of Maharani Laxmi Bai Medical College, Jhansi, for ≥72 h. Severity scores were calculated within 24 h of admission using standard formulas. Outcomes, including mortality, ICU stay, and ventilator requirement, were analyzed using receiver operating characteristic curves, correlation analysis, and logistic regression through Statistical Package for the Social Sciences v21. Results: MPM III (day 3) exhibited the highest predictive accuracy with an area under the curve of 0.88, sensitivity 90.1%, and specificity 84.6%. SOFA, MODS, and LODS also significantly differentiated survivors from non-survivors (P<0.001). Strong correlations were found between higher scores and ICU mortality, ventilator need, and longer ICU stays. Multivariate analysis identified age >60, APACHE II, SOFA, MODS, and MPM III as independent predictors of mortality. Conclusion: MPM III (day 3) is the most reliable severity scoring system for predicting ICU mortality in this setting. APACHE II and SOFA also remain clinically valuable for critical care prognostication.