Abstract
BACKGROUND: This study aims to develop a predictive model for risk factors associated with dysphagia in patients with craniocerebral injury and to propose targeted nursing and rehabilitation interventions based on this model to improve patient outcomes. METHODS: A retrospective analysis was conducted on clinical data from 150 patients with craniocerebral injury admitted between June 2022 and June 2024. Patients were divided into dysphagia (n = 62) and control (n = 88) groups based on the presence or absence of dysphagia. Univariate analysis and binary logistic regression were performed to identify independent risk factors for dysphagia, based on which a nomogram prediction model was subsequently constructed. RESULTS: Binary logistic regression identified vagus nerve injury, tracheotomy/cannulation, mechanical ventilation, and severe aphasia as independent risk factors for dysphagia, while a higher Glasgow Coma Scale (GCS) score served as a protective factor. The nomogram model based on these variables demonstrated good predictive performance, with an AUC of 0.877 (95% CI: 0.823-0.931) as validated by internal bootstrap resampling. Decision curve analysis showed no significant difference between predicted and observed outcomes (X(2) = 5.6728, p = 0.6838), and the absolute error between predicted and actual values was 0.038, indicating strong clinical utility. CONCLUSION: Vagus nerve injury, tracheotomy/cannulation, mechanical ventilation, severe aphasia, and GCS score are independent factors influencing the risk of dysphagia in craniocerebral injury patients. The developed predictive model demonstrates high accuracy and may provide a valuable reference for optimizing preventive nursing strategies and reducing the incidence of dysphagia.