ICU Mortality Prediction

Uses

Benefits

Comparisons

Background

Highlighted claims

• Non-survivors had significantly higher SHR but significantly lower HGI compared to survivors at 360 days. — Predictive value of stress hyperglycaemia ratio and haemoglobin glycation index for mortality risks in critically ill patients: a comparative retrospective analysis of the MIMIC-IV database using machine learning-based predictive modelling
• Non-survivors were older, had higher heart rates, longer ICU and hospital lengths of stay, higher OASIS scores, and greater vasopressin use than survivors. — Predictive value of stress hyperglycaemia ratio and haemoglobin glycation index for mortality risks in critically ill patients: a comparative retrospective analysis of the MIMIC-IV database using machine learning-based predictive modelling
• Both SHR and HGI require only two routinely collected laboratory values, making them straightforward to compute at the bedside. — Predictive value of stress hyperglycaemia ratio and haemoglobin glycation index for mortality risks in critically ill patients: a comparative retrospective analysis of the MIMIC-IV database using machine learning-based predictive modelling
• No prior study had directly compared both SHR and HGI in a general critically ill population before this work. — Predictive value of stress hyperglycaemia ratio and haemoglobin glycation index for mortality risks in critically ill patients: a comparative retrospective analysis of the MIMIC-IV database using machine learning-based predictive modelling
• Incorporating SHR into standard severity assessment alongside OASIS, BUN, age, and vasopressin use measurably improves 360-day mortality prediction. — Predictive value of stress hyperglycaemia ratio and haemoglobin glycation index for mortality risks in critically ill patients: a comparative retrospective analysis of the MIMIC-IV database using machine learning-based predictive modelling