ABSTRACT Objective The aim of this study was to develop a mortality risk score in the intensive care units of referral hospitals in Bahir Dar City. Methods The study included 852 participants who were admitted from January 1, 2019 to December 31, 2021. We used EpiData version 3.1 for data entry and R‐software for analysis. The mortality rate among participants was 35.9%. Multivariable logistic regression was employed to identify the independent prognostic determinants. Using beta‐coefficients, we developed and validated a prognostic model. Then a mortality risk score was determined based on the value of each prognostic determinant variable. Results Age, sex, health insurance user status, respiratory rate, temperature, mean arterial pressure, Glasgow Coma Scale, WBC count, sepsis, ARDS, organ‐insufficiency, mechanical ventilation, and vasopressor were independent prognostic determinants. Based on the prognostic determinants, we developed an easily applicable mortality risk score model. The model had a discrimination performance of AUC 0.90 (95% confidence interval of 0.88–0.92) and a calibration p value of 0.69. Conclusion The prognostic determinants identified in this study are easily accessible and easy to capture in routine clinical settings. As a result, the developed model has the potential to be effectively applied in low‐income countries where resources may be limited. Implication for Clinical Practice The model can help healthcare providers in low‐income settings to identify high‐risk patients and develop appropriate interventions to improve patient outcomes.
Wubie et al. (Wed,) studied this question.