Risk Prediction Model for Frailty in Older Chinese Patients With Type 2 Diabetes Mellitus: A Systematic Review.

AIMS: To systematically evaluate frailty risk prediction models for older patients with diabetes in China, providing a reference for healthcare professionals in selecting or developing appropriate frailty risk prediction models for older patients with diabetes and offering evidence for the formulation of intervention strategies. METHODS: A systematic search was conducted in PubMed, the Cochrane Library, Medline, Embase, Web of Science, China Biomedical Literature Database, China National Knowledge Infrastructure (CNKI), VIP Database, and Wanfang Database for studies related to frailty risk prediction models for older patients with diabetes. The search period covered from database inception to November 15, 2025. Data extraction and quality assessment were independently performed by two researchers using the Prediction Model Risk of Bias Assessment Tool (PROBAST) and a data extraction form. RESULTS: A total of 14 studies involving 25 models were included, with outcome event incidence rates ranging from 10.1% to 51.2%. The area under the receiver operating characteristic curve (AUC) of the models ranged from 0.703 to 0.975. All 14 studies showed good overall applicability but exhibited a high risk of bias. High-frequency predictors included age, polypharmacy, nutritional status, glycated hemoglobin, and ADL score. CONCLUSIONS: Frailty risk prediction models for older patients with diabetes in China demonstrate good discriminative ability and applicability, but have significant methodological flaws and a high risk of bias. Future studies should strictly follow reporting guidelines for risk prediction models to develop and evaluate frailty risk prediction models for older patients with diabetes in China, and validate their feasibility in clinical practice to provide high-quality evidence for clinical decision-making.