A decade-long real-world cohort (2016–2025): development of an individualized risk-stratification nomogram and evaluation of clinical utility for recurrent respiratory tract infections in children

Background Recurrent respiratory tract infections (RRTIs) are common in children and are associated with substantial healthcare costs and reduced quality of life. A practical tool for individualized risk assessment may facilitate the early identification of high-risk children and support targeted prevention. Methods In this 10-year real-world cohort study (2016–2025), eligible children were screened and followed according to predefined criteria. After exclusions, 6,026 children were included, of whom 1,227 (20.4%) developed RRTIs. The cohort was randomly divided into a training set ( n = 4,219; events = 859) and a testing set ( n = 1,807; events = 368). Multivariable logistic regression was used in the training set to identify independent predictors and develop a customized nomogram. Bootstrap resampling was used for internal validation, and model performance was assessed in terms of discrimination, calibration, and clinical utility using decision-curve analysis. Results Multivariable analysis identified history of allergy (OR 5.187), history of asthma (OR 2.522), lower vitamin A level (OR 0.458 per 0.1 mg/L increase), lower vitamin D level (OR 0.556 per 10 ng/mL increase), lower birth weight (OR 0.283 per kg increase), passive smoking exposure (OR 2.061), and lower hemoglobin level (OR 0.425 per 10 g/L increase) as independent predictors of RRTI (all P 0.001). The nomogram demonstrated good discrimination and calibration in both the training and testing sets. Decision-curve analysis supported its potential clinical utility by showing a favorable net benefit across clinically relevant threshold probabilities. Risk stratification based on predicted-probability cutoffs further separated children into low-, intermediate-, and high-risk groups with clearly different observed event rates. Conclusion We developed and validated a nomogram based on routinely available clinical history and nutritional indicators to predict individualized RRTI risk in children. This tool demonstrated robust predictive performance and may support early risk stratification and preventive decision-making in pediatric practice.