Construction and validation of early warning model of retinopathy risk in patients with type 2 diabetes
Objective To explore the construction and verification of a risk warning model for retinopathy in patients with type 2 diabetes mellitus(T2DM).Methods A total of 370 patients with T2DM admitted to the Second People's Hospital of Qingyang City from June 2021 to December 2022 were retrospectively selected.According to the allocation ratio of 7∶3,they were divided into the modeling group(259 cases)and the validation group(111 cases).The baseline clinical data of patients with T2DM in the modeling group were collected,and 259 patients were divided into the retinopathy group(78 cases)and the non-retinopathy group(181 cases)according to whether retinopathy occurred.Univariate analysis was used to analyze the influencing factors of retinopathy in patients with T2DM,multivariate logistic regression was used to analyze the risk factors of retinopathy in patients with T2DM,and a nomogram for retinopathy in patients with T2DM was constructed.The verification of the prediction model was completed through the data collection of the validation group.Results The incidence of patients with T2DM retinopathy was 30.12％(78/259).Single factor analysis showed that patients with age,duration,FBG,HbA1c levels,and UACR were the influence factors of T2DM retinopathy(P＜0.05).Multivariate Logistic regression analysis showed that age,disease duration ≥ 5 years,FBG level,HbA1c level and UACR were risk factors for retinopathy in patients with T2DM(P＜0.05).The area under the ROC curve(AUC)of the Nomogram model was 0.912(95％CI:0.875-0.949),and the discrimination was good.The best cut-off value was 0.587,the maximum Youden value was 0.683,the sensitivity was 0.821,and the specificity was 0.862.The theoretical and actual values of the calibration curve in the validation group and the modeling group were in good agreement.Conclusion The prediction model of retinopathy in patients with T2DM can quantify the risk of diabetic retinopathy,and can rely on scientific and effective risk factor assessment to provide personalized intervention measures to reduce the risk of diabetic retinopathy.
Type 2 diabetes mellitusRetinopathyPrediction modelFactor analysisNomogram
万方数据
维普
