Analysis of risk factors of elderly patients with type 2 diabetes complicatedwith frailty and establishment of prediction model

[Objective] To analyze the risk factors of elderly patients with type 2 diabetes complicated with frailty, and construct a nomogram. [Methods] A total of 152 elderly type 2 diabetes patients admitted to General Hospital of Nothern Theater Command of the People’s Liberation Army of China from November 2023 to January 2024 were selected, including 106 in the training set and 46 in the validation set. Logistic regression model analysis was applied to screen for risk factors of frailty in elderly type 2 diabetes patients; R software was applied to draw nomogram, ROC curve, calibration curve and Hosmer Lemeshow goodness of fit test were used to assess nomogram performance. [Results] Among the 106 elderly type 2 diabetes patients, 33 cases (31.13%) had frailty, and 73 cases (68.87%) did not have frailty. The results of logistic regression analysis showed that gender is female, stroke, depressive symptoms and high risk of falls were independent risk factors for comorbid frailty in elderly type 2 diabetes patients (P<0.05). The evaluation of training set discrimination: the area under the ROC curve was 0.840 (95% CI: 0.765 - 0.915). Consistency of training set: the slope of the calibration curve was close to 1, and the Hosmer Lemeshow goodness of fit test shows good consistency between the predicted probability of the nomogram and the actual observed results(P<0.05). Verification of cluster discrimination evaluation: the area under the ROC curve is 0.933 (95% CI: 0.864 -1.000). Verification set consistency evaluation: the slope of the calibration curve is close to 1, and the Hosmer-Lemeshow goodness of fit test shows good consistency between the predicted probability of the nomogram and the actual observed results. [Conclusion] Elderly type 2 diabetes patients are prone to comorbid frailty, with gender is female, stroke, depressive symptoms and high risk of falls are associated with comorbid frailty in elderly type 2 diabetes patients, and a nomogram constructed based on the above four independent risk factors has good predictive efficacy for comorbid frailty in elderly type 2 diabetes patients.