Biological age models based on a healthy Han Chinese population. (April 2023)

• Record Type: Journal Article
• Authors: Li, Zhe; Zhang, Weiguang; Duan, Yuting; Niu, Yue; He, Yan; Chen, Yizhi; Liu, Xiaomin; Dong, Zheyi; Zheng, Ying; Chen, Xizhao; Feng, Zhe; Wang, Yong; Zhao, Delong; Sun, Xuefeng; Cai, Guangyan; Jiang, Hongwei; Chen, Xiangmei
• Is Part Of: Archives of gerontology and geriatrics. Volume 107(2023)
• Publication Date: 2023-04
• Subjects: Biological age -- Chronological age -- Aging markers -- Age -- Healthy -- Human; Aging -- Periodicals; Geriatrics -- Periodicals; Gerontology -- Periodicals; Electronic journals; 305.26
• DOI: 10.1016/j.archger.2022.104905
• Languages: English
• ISSNs: 0167-4943

Abstract

Highlights: The study subjects were healthy aging adults, avoiding the interference of disease factors. Nine aging markers were successfully selected. Eight biological age models were successfully constructed and internally and externally validated. By comparison, the KDM2 model was found to be the most accurate in this study. Abstract: Background: Biological age (BA) may reflect the actual aging state in humans better than chronological age (CA). The study aimed to construct BA models suitable for the Chinese Han population by selecting appropriate aging markers and evaluation methods. Methods: A total of 1207 individuals (21∼91 years) from the Han Chinese population in Beijing were examined for essential organ functions, and 156 cardiovascular, pulmonary function, and atherosclerotic indices and clinical and genetic factors were used as candidate markers of aging. BA models were constructed using multiple linear regression (MLR), principal component analysis (PCA), and the Klemera and Doubal method (KDM). Models were internally and externally validated using cross-validation and disease populations. Results: Nine aging markers were selected. Two MLR, three PCA, and three KDM models were successfully constructed. External validation showed that the difference between CA and BA was most significant in the PCA3 and KDM2 models, while there was no significant difference in the MLR1 and MLR2 models; the fitted lines for BA in the disease population were higher than those in the healthy population in the MLR1, MLR2, KDM1, and KDM2 models, while the other models showed the opposite. Conclusions: Based on a healthy population in Beijing, nine markers representing multiple organ/system functions were screened from the candidate markers, eight methods were successfully used to construct BA models, and the KDM2 model was found to potentially be more appropriate for assessing BA in the Chinese Han population.