The precision of mortality prediction by a novel checklist (M score) compared with well-known scoring systems in non-trauma patients in intensive care unit

Abstract

Background: It seem necessary to assess the precision of scoring systems for estimating disease severity and prognosis of the patients in intensive care units (ICUs). In this study, we aimed to compare the predictive accuracy of a novel checklist (M score) with well-known scoring systems in non-trauma patients in ICU. Methods: Throughout a cross-sectional study in 2018, 835 non-trauma patients admitted in ICU with the age of more than 16 years were included. Mortality prediction was assessed using Acute Physiology and Chronic Health Evaluation II (APACHE II), Sequential Organ Failure Assessment (SOFA), and M score at the first and last day of their ICU stay. Using logistic regression test, receiver operation curve, Pearson's regression, chi-square, and independent samples t tests, the data from these three scoring systems were analyzed, and the sensitivity and specificity of systems were calculated. Findings: The cut-off point for predicting mortality was 13.5 for APACHE II, 15.5 for M score, and 6.5 for SOFA. The probability of death increased by increasing in any of studied scores. Conclusion: M score, which has been prepared for accurate easement of clinical and paraclinical status of patients by nurses and physicians, seems to be a good corrival for well-known scoring systems. © 2019 Isfahan University of Medical Sciences(IUMS). All rights reserved.