Risk Prediction Models for Frailty in Maintenance Hemodialysis Patients: A Systematic Review and Meta‐Analysis

OBJECTIVE: The aim is to systematically evaluate the frailty risk prediction models for Chinese maintenance hemodialysis patients and to provide a reference for the construction and optimization of such models. METHODS: Relevant studies on prediction models for frailty in Chinese maintenance hemodialysis patients were retrieved from PubMed, Embase, Web of Science, Cochrane Library, CINAHL, China Biomedical Literature Database, China National Knowledge Infrastructure, Wanfang Database, and VIP Database. The search period was from the inception of each database to October 23, 2024. Two researchers independently screened the literature, extracted data, and assessed the quality of the included models using the Prediction model Risk of Bias Assessment Tool (PROBAST). A meta-analysis of predictors was performed using Stata 18.0 software. RESULTS: A total of 13 studies from China were included, which developed 15 distinct prediction models. These models involved 4341 patients. The area under the receiver operating characteristic curve (AUC) for these models ranged from 0.722 to 0.998, indicating good predictive performance (AUC > 0.7). However, the overall risk of bias was high, primarily in the domain of data analysis. The meta-analysis identified the following significant predictors of frailty (p < 0.05): age (OR = 1.14, 95% CI 1.05-1.23), serum albumin (OR = 0.66, 95% CI 0.52-0.83), exercise (OR = 0.50, 95% CI 0.49-0.63), comorbidity (OR = 1.70, 95% CI 1.47-1.97), nutritional score (OR = 3.64, 95% CI 1.53-8.67), ADL score (OR = 0.77, 95% CI 0.68-0.88), female sex (OR = 6.24, 95% CI 1.97-19.80), and depression (OR = 1.26, 95% CI 1.04-1.53). CONCLUSION: Our results indicate the incidence of frailty among maintenance hemodialysis patients in China is as high as 39%. The identified predictors-advanced age, hypoalbuminemia, physical inactivity, comorbidities, poor nutritional status, impaired activities of daily living, female sex, and depression-form the basis for developing targeted preventive measures for frail patients on maintenance hemodialysis. The prediction model of frailty risk in maintenance hemodialysis patients in China is still in its infancy. Future research can refer to the model construction method of this study and the common predictors integrated by meta-analysis and select appropriate methods to develop and verify the frailty prediction model in combination with clinical practice. Targeted preventive measures should be given to maintenance hemodialysis patients with high risk in the early stage.