Development and validation of a risk prediction model for mild cognitive impairment in elderly patients with type 2 diabetes mellitus

• Published in: Geriatric nursing (New York) Vol. 58; pp. 119 - 126
• Main Authors: Yu, Qian, Jiang, Xing, Yan, Jiarong, Yu, Hongyu
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.07.2024
• Subjects: Aged; Cognitive dysfunction; Diabetes mellitus; Nomogram; Prediction model; Aged; Diabetes mellitus; Nomogram; Cognitive dysfunction; Prediction model
• ISSN: 0197-4572; 1528-3984; 1528-3984
• DOI: 10.1016/j.gerinurse.2024.05.018

•Nearly a quarter of individuals with pre-T2DM experience mild cognitive impairment.•Age, living alone, smoking, physical activity, social support, depression, and HBA1c status were found to be associated with mild cognitive impairment.•The nomogram provides a clinical basis for detecting mild cognitive impairment in elderly patients with type 2 diabetes mellitus. The prevalence of mild cognitive impairment (MCI) is steadily increasing among elderly people with type 2 diabetes (T2DM). This study aimed to create and validate a predictive model based on a nomogram. This cross-sectional study collected sociodemographic characteristics, T2DM-related factors, depression, and levels of social support from 530 older adults with T2DM. We used LASSO regression and multifactorial logistic regression to determine the predictors of the model. The performance of the nomogram was evaluated using calibration curves, receiver operating characteristics (ROC), and decision curve analysis (DCA). The nomogram comprised age, smoking, physical activity, social support, depression, living alone, and glycosylated hemoglobin. The AUC for the training and validation sets were 0.914 and 0.859. The DCA showed good clinical applicability. This predictive nomogram has satisfactory accuracy and discrimination. Therefore, the nomogram can be intuitively and easily used to detect MCI in elderly adults with T2DM.