Development and Validation of a Risk Prediction Model for Depression in Elderly Chinese Patients on Maintenance Hemodialysis: A Multicenter Study

Background Depression is the most prevalent psychological condition among elderly patients on maintenance hemodialysis, significantly impairing quality of life and increasing readmission and mortality risks. Methods This study recruited 871 elderly patients undergoing hemodialysis from nine tertiary hospitals in China between November 2023 and February 2024. Depression symptoms were assessed using the Geriatric Depression Scale (GDS‐15), with a score of ≥ 8 indicating clinically significant depressive symptoms. Participants were divided into depressed ( n = 333) and nondepressed ( n = 538) groups. Multivariable logistic regression analysis was used to identify independent risk factors and develop a risk prediction model presented as a nomogram, with internal validation performed. External validation was conducted in an independent cohort of 219 elderly patients undergoing maintenance hemodialysis recruited from three additional hospitals in Chengdu between March and April 2024. Results The prevalence of depressive symptoms among elderly maintenance hemodialysis patients was 38.2%. Logistic regression identified education level, vision impairment, frailty, cognitive impairment, malnutrition, low activities of daily living, and poor social support as independent risk factors ( p < 0.05). Both internal and external validation of the model demonstrated a receiver operating characteristic (ROC) curve area under the curve (AUC) greater than 0.80, indicating good model discrimination, with calibration and clinical decision analyses confirming its clinical utility. Conclusions The high prevalence of depressive symptoms in this population is linked to specific risk factors. The nomogram provides valuable support for identification of high‐risk patients in clinical practice.