Abstract
BACKGROUND: Dysphagia is a common complication in elderly patients with frailty, affecting their prognosis and quality of life. Constructing a risk prediction model can help with early screening and intervention. OBJECTIVE: To investigate the current status of dysphagia in hospitalized elderly patients with frailty, analyze its influencing factors, and construct a risk prediction model for dysphagia in hospitalized elderly patients with frailty. METHODS: A total of 300 hospitalized elderly patients with frailty were selected as research subjects using a convenience sampling method from May to December 2024 in a tertiary general hospital in Mianyang. The survey tools included the General Information Questionnaire, Mini Nutritional Assessment-Short Form (MNA-SF), Geriatric Self-Efficacy Scale for Oral Health (GSEOH), Geriatric Oral Health Assessment Index (GOHAI), and 5-Item Geriatric Depression Scale (GDS-15). Data were analyzed using SPSS 26.0 software, and variable selection was conducted using the backward LR method to construct the risk prediction model. RESULTS: A total of 300 questionnaires were distributed, and 287 valid questionnaires were retrieved, with an effective recovery rate of 95.7%. Among the 287 patients, 103 cases (35.9%) were identified with dysphagia. Among the 202 patients with a history of choking, 80 cases (39.6%) were identified as having swallowing disorders. In contrast, among the 85 patients without a history of choking, 23 cases (27.1%) had swallowing disorders. The difference was statistically significant (χ (2) = 4.092, p = 0.043). Logistic regression analysis showed that age, history of coughing, polypharmacy, malnutrition, oral health-related self-efficacy, and oral health assessment index were risk factors for dysphagia in elderly patients with frailty (p < 0.05). The constructed risk prediction model was: Logit p = 0.770 × Age + 0.919 × Polypharmacy + 1.009 × History of Coughing + 1.208 × Malnutrition - 0.113 × Oral Health-Related Self-Efficacy - 0.262 × Oral Health Assessment Index + 10.200. The Hosmer-Lemeshow goodness-of-fit test indicated no statistically significant difference between the model's predictions and actual outcomes (χ (2) = 6.939, p = 0.543, p > 0.05). The area under the ROC curve (AUC) was 0.875, with a sensitivity of 0.631 and a specificity of 0.891. CONCLUSION: The incidence of dysphagia in hospitalized elderly patients with frailty is relatively high. The main influencing factors include age, history of coughing, polypharmacy, malnutrition, oral health assessment index, and oral health-related self-efficacy. Healthcare professionals should enhance their awareness of risk warning, conduct early screening, and implement preventive measures. The constructed risk prediction model demonstrates high calibration and discrimination abilities, providing a valuable reference for the early detection, prevention, and intervention of dysphagia in hospitalized elderly patients with frailty.