Development and validation of a nomogram model for individualized prediction of hypertension risk in patients with type 2 diabetes mellitus

Abstract

Type 2 diabetes mellitus (T2DM) with hypertension (DH) is the most common diabetic comorbidity. Patients with DH have significantly higher rates of cardiovascular disease morbidity and mortality. The objective of this study was to develop and validate a nomogram model for the prediction of an individual's risk of developing DH. A total of 706 T2DM patients who met the criteria were selected and divided into a training set (n = 521) and a validation set (n = 185) according to the discharge time of patients. By using multivariate logistic regression analysis and stepwise regression, the DH nomogram prediction model was created. Calibration curves were used to evaluate the model's accuracy, while decision curve analysis (DCA) and receiver operating characteristic (ROC) curves were used to evaluate the model's clinical applicability and discriminatory power. Age, body mass index (BMI), diabetic nephropathy (DN), and diabetic retinopathy (DR) were all independent risk factors for DH (P < 0.05). Based on independent risk factors identified by multivariate logistic regression, the nomogram model was created. The model produces accurate predictions. If the total nomogram score is greater than 120, there is a 90% or higher chance of developing DH. In the training and validation sets, the model's ROC curves are 0.762 (95% CI 0.720–0.803) and 0.700 (95% CI 0.623–0.777), respectively. The calibration curve demonstrates that there is good agreement between the model’s predictions and the actual outcomes. The decision curve analysis findings demonstrated that the nomogram model was clinically helpful throughout a broad threshold probability range. The DH risk prediction nomogram model constructed in this study can help clinicians identify individuals at high risk for DH at an early stage, which is a guideline for personalized prevention and treatments.