Construction of a new biological age model based on NHANES database and its predictive role for mortality outcomes
Objective To develop a new biological age model based on the National Health and Nutrition Examination Survey (NHANES) database and evaluate its predictive performance for mortality outcomes. Methods After excluding pregnant women and those with incomplete clinical data, a total of 8 234 particip...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
Editorial Office of Journal of Army Medical University
2025-04-01
|
| Series: | 陆军军医大学学报 |
| Subjects: | |
| Online Access: | https://aammt.tmmu.edu.cn/html/202410082.html |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849712487688044544 |
|---|---|
| author | YE Tingting GAO Pan |
| author_facet | YE Tingting GAO Pan |
| author_sort | YE Tingting |
| collection | DOAJ |
| description | Objective To develop a new biological age model based on the National Health and Nutrition Examination Survey (NHANES) database and evaluate its predictive performance for mortality outcomes. Methods After excluding pregnant women and those with incomplete clinical data, a total of 8 234 participants aged 20~79 years from NHANES 2007-2010 were included and assigned into a testing set, and another 17 522 non-pregnant participants （17~79 years old） subjected from NHANESⅢ were enrolled and served as a training set to construct the new biological age (new Bioage) model. Multiple linear regression was used to construct models for new Bioage and age-adjusted biological age (new-agefixed Bioage), and other types of biological ages were calculated. Pearson correlation analysis was performed to assess the correlation between biological age and chronological age. Based on new-agefixed Bioage, biological age acceleration was calculated, and then according to the results, the participants were divided into accelerated biological aging group (biological age acceleration ≥0, n=3 884) and decelerated biological aging group (biological age acceleration <0, n=4 350). Weighted data comparisons were conducted using the survey package in R, and Cox regression analysis was used to assess the impact of biological age acceleration on the mortality of the participants. Receiver operating characteristic (ROC) curve was plotted to determine the effect of new-agefixed Bioage and biological age acceleration on the area under the curve (AUC) for the mortality. Results The new-agefixed Bioage model, composed of 9 variables, had the highest AUC value (AUC=0.889 2, P<0.001) in predicting mortality outcomes and showed high correlation with chronological age and other biological ages. The decelerated biological aging group had significantly larger proportions of males, whites, education level of high school or higher, middle to high household income, married (or with a partner), engaging in physical activity, having private insurance, and cancer (P<0.05), lower BMI, slower biological age acceleration level, and less proportions of smokers, hypertension, diabetes, cardiovascular disease, and chronic obstructive pulmonary disease when compared with the accelerated biological aging group (P<0.05). Cox regression analysis indicated that, compared to the decelerated biological aging group, the accelerated biological aging group had a significantly increased risk of mortality in Model3 [adjust for age group, gender, race, BMI, education level, household income, married (or with a partner), smoking status, heavy alcohol consumption, insurance status, physical activity, hypertension, diabetes, cardiovascular disease, chronic obstructive pulmonary disease, and cancer](HR=1.62, 95% CI: 1.28-2.06, P<0.001). ROC curve analysis revealed that new-agefixed Bioage significantly increased the AUC values (AUC=0.781, P<0.001; AUC=0.731, P<0.001) for middle-aged and elderly populations, while biological age acceleration significantly improved the AUC values (AUC=0.756, P<0.001) for the middle-aged population. Conclusion The new-agefixed Bioage model demonstrates high correlation with chronological age and other biological ages, and has good predictive performance for mortality outcomes, particularly in middle-aged and elderly populations.
|
| format | Article |
| id | doaj-art-1eaeecebd0f74f72a136de42ea989d65 |
| institution | DOAJ |
| issn | 2097-0927 |
| language | zho |
| publishDate | 2025-04-01 |
| publisher | Editorial Office of Journal of Army Medical University |
| record_format | Article |
| series | 陆军军医大学学报 |
| spelling | doaj-art-1eaeecebd0f74f72a136de42ea989d652025-08-20T03:14:15ZzhoEditorial Office of Journal of Army Medical University陆军军医大学学报2097-09272025-04-0147887688410.16016/j.2097-0927.202410082Construction of a new biological age model based on NHANES database and its predictive role for mortality outcomesYE Tingting0GAO Pan1Department of Geriatric and Special Medical Services， First Affiliated Hospital， Army Medical University （Third Military Medical University）， Chongqing， ChinaDepartment of Geriatric and Special Medical Services， First Affiliated Hospital， Army Medical University （Third Military Medical University）， Chongqing， ChinaObjective To develop a new biological age model based on the National Health and Nutrition Examination Survey (NHANES) database and evaluate its predictive performance for mortality outcomes. Methods After excluding pregnant women and those with incomplete clinical data, a total of 8 234 participants aged 20~79 years from NHANES 2007-2010 were included and assigned into a testing set, and another 17 522 non-pregnant participants （17~79 years old） subjected from NHANESⅢ were enrolled and served as a training set to construct the new biological age (new Bioage) model. Multiple linear regression was used to construct models for new Bioage and age-adjusted biological age (new-agefixed Bioage), and other types of biological ages were calculated. Pearson correlation analysis was performed to assess the correlation between biological age and chronological age. Based on new-agefixed Bioage, biological age acceleration was calculated, and then according to the results, the participants were divided into accelerated biological aging group (biological age acceleration ≥0, n=3 884) and decelerated biological aging group (biological age acceleration <0, n=4 350). Weighted data comparisons were conducted using the survey package in R, and Cox regression analysis was used to assess the impact of biological age acceleration on the mortality of the participants. Receiver operating characteristic (ROC) curve was plotted to determine the effect of new-agefixed Bioage and biological age acceleration on the area under the curve (AUC) for the mortality. Results The new-agefixed Bioage model, composed of 9 variables, had the highest AUC value (AUC=0.889 2, P<0.001) in predicting mortality outcomes and showed high correlation with chronological age and other biological ages. The decelerated biological aging group had significantly larger proportions of males, whites, education level of high school or higher, middle to high household income, married (or with a partner), engaging in physical activity, having private insurance, and cancer (P<0.05), lower BMI, slower biological age acceleration level, and less proportions of smokers, hypertension, diabetes, cardiovascular disease, and chronic obstructive pulmonary disease when compared with the accelerated biological aging group (P<0.05). Cox regression analysis indicated that, compared to the decelerated biological aging group, the accelerated biological aging group had a significantly increased risk of mortality in Model3 [adjust for age group, gender, race, BMI, education level, household income, married (or with a partner), smoking status, heavy alcohol consumption, insurance status, physical activity, hypertension, diabetes, cardiovascular disease, chronic obstructive pulmonary disease, and cancer](HR=1.62, 95% CI: 1.28-2.06, P<0.001). ROC curve analysis revealed that new-agefixed Bioage significantly increased the AUC values (AUC=0.781, P<0.001; AUC=0.731, P<0.001) for middle-aged and elderly populations, while biological age acceleration significantly improved the AUC values (AUC=0.756, P<0.001) for the middle-aged population. Conclusion The new-agefixed Bioage model demonstrates high correlation with chronological age and other biological ages, and has good predictive performance for mortality outcomes, particularly in middle-aged and elderly populations. https://aammt.tmmu.edu.cn/html/202410082.htmlnational health and nutrition examination survey databasebioagebioage accelerationphenoage |
| spellingShingle | YE Tingting GAO Pan Construction of a new biological age model based on NHANES database and its predictive role for mortality outcomes 陆军军医大学学报 national health and nutrition examination survey database bioage bioage acceleration phenoage |
| title | Construction of a new biological age model based on NHANES database and its predictive role for mortality outcomes |
| title_full | Construction of a new biological age model based on NHANES database and its predictive role for mortality outcomes |
| title_fullStr | Construction of a new biological age model based on NHANES database and its predictive role for mortality outcomes |
| title_full_unstemmed | Construction of a new biological age model based on NHANES database and its predictive role for mortality outcomes |
| title_short | Construction of a new biological age model based on NHANES database and its predictive role for mortality outcomes |
| title_sort | construction of a new biological age model based on nhanes database and its predictive role for mortality outcomes |
| topic | national health and nutrition examination survey database bioage bioage acceleration phenoage |
| url | https://aammt.tmmu.edu.cn/html/202410082.html |
| work_keys_str_mv | AT yetingting constructionofanewbiologicalagemodelbasedonnhanesdatabaseanditspredictiveroleformortalityoutcomes AT gaopan constructionofanewbiologicalagemodelbasedonnhanesdatabaseanditspredictiveroleformortalityoutcomes |