Construction of a nomogram prediction model for microvascular complications in patients with type 2 diabetes mellitus under the Metabolic Management Center mode
Objective To analyze the factors affecting microvascular complications in patients with type 2 diabetes mellitus(T2DM)under the Metabolic Management Center(MMC)mode,and to construct a nomogram prediction model.Methods Patients with T2DM from Endocrinology Department of Tangshan Central Hospital from June 2019 to December 2020 were selected as the subjects.After the enrollment,they were provided with comprehensive management through the MMC platform,and all the patients were followed up for 3 years.The T2DM patients'basic data and status of coexistence of microvascular complications like retinopathy,nephropathy and neuropathy were collected.Lasso regression and multifactor logistic regression were used to screen the factors influencing microvascular complications in the T2DM patients,and construct a nomogram prediction model.The receiver operating characteristic(ROC)curve,Bootstrap calibration curve and decision curve were used to evaluate the prediction ability of the nomogram model for microvascular complications in the T2DM patients.Results A total of 300 T2DM patients were enrolled into this study,and 14 were lost to follow-up,eventually enrolling 286 patients.Among the 286 T2DM patients,123(43.01％)had microvascular complications.The results of multifactor logistic regression displayed that body mass index(OR=1.167,95％CI:1.062-1.283),diabetes duration(OR=1.328,95％CI:1.130-1.562),total bilirubin(OR=0.937,95％CI:0.899-0.977),blood urea nitrogen(OR=1.300,95％CI:1.086-1.555)and systolic blood pressure(OR=1.029,95％CI:1.015-1.044)were the factors influencing microvascular complications in the T2DM patients(P＜0.05).The ROC curve analysis revealed that the area under the curve of the model was 0.779(95％CI:0.726-0.826),the specificity 0.767,and the sensitivity 0.675.The Hosmer-Lemeshow goodness-of-fit test also demonstrated that the nomogram model had a good fit(x2=10.386,P=0.239).The decision curve suggested that when the threshold probability was in the range of 0.06-0.68,the net benefit of using the nomogram to predict the risk of microvascular complications in the T2DM patients was higher.Conclusion Based on 5 conventional prediction indicators,this study constructs a nomogram model for forecasting microvascular complications in the T2DM patients under the MMC mode,and the model has a good predictive accuracy.The study results still need external validation.
NETL
NSTL
万方数据
