Department of Vascular Surgery Xuanwu Hospital Capital Medical University, 45 Changchun Street, Xicheng District, Beijing, China.
J Diabetes Res. 2024 Aug 19;2024:5583707. doi: 10.1155/2024/5583707. eCollection 2024.
The purpose of this study is to establish a predictive model of cognitive impairment in elderly people with diabetes. We analyzed a total of 878 elderly patients with diabetes who were part of the National Health and Nutrition Examination Survey (NHANES) from 2011 to 2014. The data were randomly divided into training and validation cohorts at a ratio of 6:4. The least absolute shrinkage and selection operator (LASSO) logistic regression analysis to identify independent risk factors and construct a prediction nomogram for cognitive impairment. The performance of the nomogram was assessed using receiver operating characteristic (ROC) curve and calibration curve. Decision curve analysis (DCA) was performed to evaluate the clinical utility of the nomogram. LASSO logistic regression was used to screen eight variables, age, race, education, poverty income ratio (PIR), aspartate aminotransferase (AST), blood urea nitrogen (BUN), serum uric acid (SUA), and heart failure (HF). A nomogram model was built based on these predictors. The ROC analysis of our training set yielded an area under the curve (AUC) of 0.786, while the validation set showed an AUC of 0.777. The calibration curve demonstrated a good fit between the two groups. Furthermore, the DCA indicated that the model has a favorable net benefit when the risk threshold exceeds 0.2. The newly developed nomogram has proved to be an important tool for accurately predicting cognitive impairment in elderly patients with diabetes, providing important information for targeted prevention and intervention measures.
本研究旨在建立一个预测糖尿病老年患者认知障碍的模型。我们分析了 2011 年至 2014 年国家健康和营养检查调查（NHANES）中总共 878 名患有糖尿病的老年患者的数据。数据被随机分为训练集和验证集，比例为 6:4。使用最小绝对收缩和选择算子（LASSO）逻辑回归分析来识别独立的危险因素，并构建认知障碍预测列线图。通过接受者操作特征（ROC）曲线和校准曲线评估列线图的性能。通过决策曲线分析（DCA）评估列线图的临床实用性。LASSO 逻辑回归用于筛选出 8 个变量，包括年龄、种族、教育、贫困收入比（PIR）、天冬氨酸氨基转移酶（AST）、血尿素氮（BUN）、血清尿酸（SUA）和心力衰竭（HF）。基于这些预测因子构建了一个列线图模型。我们的训练集的 ROC 分析得到曲线下面积（AUC）为 0.786，而验证集的 AUC 为 0.777。校准曲线表明两组之间拟合良好。此外，DCA 表明当风险阈值超过 0.2 时，该模型具有良好的净收益。新开发的列线图已被证明是准确预测糖尿病老年患者认知障碍的重要工具，为有针对性的预防和干预措施提供了重要信息。
