Abstract
OBJECTIVES: Developmental delay in preterm infants is a critical clinical issue, and early risk identification and prediction are essential. This study aims to develop and validate a predictive model for developmental delay, providing a scientific basis for clinical risk assessment and early intervention. METHODS: This study included preterm infants and their primary caregivers who were followed up at our center from May 2023 to September 2024. The samples were randomly divided into a training cohort, an internal validation cohort, and an external validation cohort. Independent risk factors for stunting were identified through univariate and multivariate logistic regression analyses, and predictive models and calibration were constructed accordingly. RESULTS: The five standardized indicators at 3, 6, 9, and 12 months for 507 preterm infants were analyzed using principal component analysis, and their developmental outcomes were grouped accordingly. Logistic regression analyses showed that gestational age, high-risk factors, knowledge of caregiving, caregiving experience, and the presence of other caregivers in the home were independent risk factors for the risk of preterm infants with stunted growth at 3, 6, 9, and 12 months. The nomogram showed the area under the receiver operating characteristic curve values of 0.743, 0.735, 0.752, and 0.774 in the training cohort; 0.855, 0.771, 0.870, and 0.786 in the internal validation cohort; 0.822, 0.804, 0.717, and 0.678 in the external validation cohort, respectively. The calibration curves, consistency index, and decision curve analysis all showed that the model was significantly better than a single indicator in predicting the risk of stunting in preterm infants. CONCLUSIONS: The stunting risk prediction model constructed in this study shows good predictive ability, which can help clinicians assess the risk of stunting in preterm infants and support the development of early intervention strategies.