Abstract
OBJECTIVE: To identify independent risk factors for protein-energy malnutrition (PEM) in children aged 8-10 years and to develop and validate a nomogram model for estimating PEM risk. METHODS: In this retrospective study, a cohort of 1,412 children from The Fifth Affiliated Hospital of Guangzhou Medical University, spanning January 2022 to December 2023, was identified. Participants were randomly classified into a training set (n=988) and a validation set (n=424). Patients in the training set were divided into normal (n=667) and PEM (n=321) groups. Data collection involved demographic, sociological, physical, and biochemical assessments. Independent risk factors for PEM were identified using univariate and multivariate logistic regression. A nomogram risk model was constructed from significant predictors, and its performance was assessed using the receiver operating characteristic (ROC) curve, calibration curve, and decision curve analysis (DCA). An independent dataset further validated the nomogram model. RESULTS: Among the 1,412 children, 497 (35.2%) had PEM, which included stunting (11.83%), underweight (11.61%), and wasting (11.76%). Multivariate analysis identified six independent risk factors for PEM: gestational age (OR (95% CI)=5.830 (3.604-9.431), P<0.001), household income (OR (95% CI)=0.383 (0.281-0.523), P<0.001), sleep duration (OR (95% CI)=1.800 (1.319-2.457), P<0.001), mood disorders (OR (95% CI)=6.924 (4.437-10.805), P<0.001), and physical activity time (OR (95% CI)=3.210 (2.342-4.400), P<0.001). The nomogram model demonstrated good predictive performance (AUC=0.803 (0.773-0.832)) and was validated well on an independent dataset (AUC=0.783 (0.739-0.828)). CONCLUSION: The study identified key independent risk factors for PEM in children and established a robust nomogram model for clinical risk assessment. The model's high predictive accuracy and clinical applicability suggest it may be a valuable tool for the early identification and intervention strategies for PEM in clinical practice.