Abstract
Purpose: Develop a diabetic nephropathy incidence risk nomogram in a Chinese population with type 2 diabetes mellitus.Results: Predictors included systolic blood pressure, diastolic blood pressure, fasting blood glucose, glycosylated hemoglobin A1c, total triglycerides, serum creatinine, blood urea nitrogen and body mass index. The model displayed medium predictive power with a C-index of 0.744 and an area under curve of 0.744. Internal verification of C-index reached 0.737. The decision curve analysis showed the risk threshold was 20%. The value of net reclassification improvement and integrated discrimination improvement were 0.131, 0.05, and that the nomogram could be applied in clinical practice.Conclusion: Diabetic nephropathy incidence risk nomogram incorporating 8 features is useful to predict diabetic nephropathy incidence risk in type 2 diabetes mellitus patients.Methods: Questionnaires, physical examinations and biochemical tests were performed on 3489 T2DM patients in six communities in Shanghai. LASSO regression was used to optimize feature selection by running cyclic coordinate descent. Logistic regression analysis was applied to build a prediction model incorporating the selected features. The C-index, calibration plot, curve analysis, forest plot, net reclassification improvement, integrated discrimination improvement and internal validation were used to validate the discrimination, calibration and clinical usefulness of the model.
Highlights
Over the past 20 years, due to the increase in the obesity rate and the prevalence of sedentary lifestyles, the number of people worldwide diagnosed with diabetes has been increasing [1], which is rapidly becoming a public health problem in both developed and developing countries [2]
All 3489 community-based patients with type 2 diabetes mellitus (T2DM) were divided into nondiabetic nephropathy (NDN) and Diabetic nephropathy (DN) groups according to the ratio of urinary microalbumin to uric creatinine (ACR), and 701 patients were diagnosed with DN
Lifestyle habits, physical examination results and biochemical test results, 19 features were reduced to 8 potential predictors on the basis of the results of 3489 patients (~2:1 ratio; Figures 1, 2) and had nonzero coefficients in the least absolute shrinkage and selection operator (LASSSO) regression model
Summary
Over the past 20 years, due to the increase in the obesity rate and the prevalence of sedentary lifestyles, the number of people worldwide diagnosed with diabetes has been increasing [1], which is rapidly becoming a public health problem in both developed and developing countries [2]. According to the report released by the International Diabetes Federation in 2015, the prevalence of diabetes among adults worldwide is 9.1% [3]. According to the latest international diagnostic criteria, 11.6% of Chinese adults had diabetes in 2010, the prevalence of prediabetes was estimated at 50.1% [12], and approximately 114 million people had type 2 diabetes mellitus (T2DM)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
