Factores predictivos de riesgo nutricional en pacientes hospitalizados en el Hospital General del Norte de Guayaquil IESS Ceibos, periodo 2024-2025.

Abstract

Introduction: Malnutrition in hospitalized patients is a common public health issue that negatively impacts clinical outcomes, increasing complications, hospital length of stay, and healthcare costs. Timely detection is essential to prevent morbidity and improve patient prognosis. Objective: To identify predictive factors of nutritional risk in hospitalized patients through the analysis of clinical, demographic, and biochemical variables. Results: A cross-sectional study was conducted involving 274 patients, using validated tools such as the NRS-2002 for risk classification. A total of 13.5% of patients were classified as high nutritional risk. Male sex, enteral nutrition, and a diagnosis of undernutrition were significantly associated with increased nutritional risk. Although some biomarkers, including C-reactive protein, LDL, and glucose, showed relevant trends, they were not statistically significant in the multivariate model. The predictive model demonstrated excellent discriminative capacity (AUC = 0.904), with an overall classification accuracy of 91.5%. Discussion and Conclusion: The proposed model proves to be a valuable tool for the early identification of hospitalized patients at high nutritional risk. It is recommended to implement predictive tools in clinical practice to optimize nutritional management and reduce related complications.