The number of elderly patients undergoing maintenance hemodialysis (MHD) in China is constantly increasing, with an upward trend. Meanwhile, the frailty of elderly patients undergoing MHD is closely associated with decreased quality of life, hospitalization, falls, disability, and mortality. To explore the influencing factors of frailty in elderly patients undergoing MHD, construct a frailty diagnosis model, and validate its efficacy, in order to provide a reference for clinical nursing practice. A total of 296 elderly patients receiving MHD treatment at 2 tertiary Class A hospitals in Hohhot from June to November 2024 were selected as research subjects. The basic inclusion criteria were patients aged ≥60 years who have undergone regular dialysis for ≥3 months. This model performed a classification diagnostic task, namely, using readily available clinical indicators to determine whether a current state of frailty exists. Patient assessments were conducted using the General Information Questionnaire, Charlson comorbidity index, Edmonton Frailty Scale, Social Support Rating Scale, Barthel index, Nutritional Risk Screening (Nutritional Risk Screening Tool 2002), Pittsburgh Sleep Quality Index, and Brief Depression Scale for the Elderly (Geriatric Depression Scale-15). Univariate analysis and Least Absolute Shrinkage and Selection Operator regression analysis were employed to identify risk factors for frailty in elderly MHD patients, followed by multivariate logistic regression modeling to establish a frailty diagnosis model. Among the 296 elderly MHD patients, 189 cases (63.9%) exhibited frailty. Least Absolute Shrinkage and Selection Operator regression analysis revealed that hemoglobin levels, serum albumin levels, serum creatinine levels, Charlson Comorbidity Index, Nutritional Risk Screening (Nutritional Risk Screening Tool 2002), Brief Depression Scale for the Elderly (Geriatric Depression Scale-15), Social Support Rating Scale, and Barthel index were independent influencing factors for frailty in elderly MHD patients. The model achieved an area under the curve of 0.899 (95% confidence interval: 0.855–0.942) with an optimal cutoff value of 0.719, demonstrating 89.6% sensitivity, 75.3% specificity, and a maximum Youden index of 0.672. External validation further confirmed the model’s robustness. The frailty diagnosis model for elderly patients undergoing MHD constructed in this study has good discriminatory efficacy and has certain clinical application value.
Zhao et al. (Fri,) studied this question.