Objective:To investigate the relationship of several adiposity-related anthropometric parameters, including BMI, waist circumference (WC), waist-to-hip ratio (WHR), body fat percentage (BFP) and indoles in plasma with the incidence of atherosclerotic cardiovascular disease (ASCVD) in adults in China.Methods:In China Kadoorie Biobank (CKB) study, blood samples were collected from 2 183 participants in the first resurvey in 2008 to detect indoles. Participants' body weight, body height, WC, hip circumference, and BFP were measured at baseline survey in 2004 and resurvey in 2008, the BMI and WHR were calculated with standardized methods. The long-term follow-up of all participants started from the completion of the resurvey in 2008 until the occurrence of incident ASCVD, death, loss to follow-up or until December 31, 2018. CKB ascertained outcome status (incident ASCVD) through death and disease registries and national health insurance databases, supplemented by active follow-up. Multivariate linear regression model was used to estimate the associations of anthropometric measurements at baseline survey and the first resurvey, and changes in these measurements with 3 indoles [indole, indole-3-acetic acid (IAA), and indole-3-propionic acid (IPA)]. Cox proportional hazard regression model was used to estimate the associations between indoles and the risk for ASCVD.Results:Anthropometric measurements at baseline survey or the first resurvey were negatively associated with plasma IPA level. The regression coefficient ( β) of baseline BMI (per 1.0 kg/m 2) with 0.1 standard deviation ( SD) IPA was -0.23 (95% CI: -0.36 - -0.10) (false discovery rate=0.004). After adjusting for baseline BMI, the β of baseline WC, WHR and BFP with 0.1 SD IPA were -0.09 (95% CI: -0.18 - -0.01), -0.12 (95% CI: -0.19 - -0.05), and -0.20 (95% CI: -0.32 - -0.08), respectively. The annual change in BMI (difference between BMI in 2008 and 2004 divided by the time interval) was associated with indole and IAA, with β of 1.40 (95% CI: 0.58 - 2.21) and -1.07 (95% CI: -1.91 - -0.23), respectively, at each 0.1 increase of SD. Over a median ( Q1, Q3) follow-up of 10.46 (10.36, 10.53) years after 2008 resurvey, 236 cases of ASCVD were recorded. IAA and IPA levels were negatively associated with the risk for ASCVD, with hazard ratios for one SD increase of IAA and IPA of 0.87 (95% CI: 0.76 - 0.99) and 0.84 (95% CI: 0.73 - 0.96), respectively. Conclusions:Our results suggested that anthropometric measurements and their changing trends affect the levels of plasma imicrobial tryptophan metabolite levels, decreased levels of IAA and IPA levels are associated with increased risk of ASCVD and indoles in plasma including IPA and IAA might be the mediating factors for adiposity-induced ASCVD.

Objective:To analyze the prospective associations between liver biomarkers and mortality among Chinese middle-aged and elderly populations and to evaluate the mortality risk predictive value.Methods:A total of 22 758 participants from the 3 rd resurvey of the China Kadoorie Biobank were included. Cox proportional hazard models were used to analyze the prospective associations of 5 liver biomarkers with mortality. These liver biomarkers included two liver imaging biomarkers (liver fat attenuation parameter, liver stiffness measurement) and three serum liver enzyme biomarkers [gamma-glutamyl transferase (GGT), ALT, and AST]. Restricted cubic spline was used to assess the nonlinear associations between biomarkers and mortality. The area used the receiver operating characteristic curve (AUC) to evaluate the predictive ability of the models after incorporating liver biomarkers into traditional prediction models for mortality. Results:The mean age of the participants was (65.2±9.1) years, with a median follow-up of 1.5 years, during which 307 deaths occurred. Compared to individuals without hepatic steatosis, those with severe hepatic steatosis had a 79% higher risk of mortality, with a HR of 1.79 (95% CI: 1.06-3.03). Compared to individuals without hepatic fibrosis, those with advanced fibrosis and cirrhosis had higher mortality risks of 48% and 91%, respectively (both P<0.05). For each standard deviation increase in GGT, the mortality risk increased by 10% ( HR=1.10, 95% CI: 1.05-1.15), with the positive association plateauing at higher GGT levels. AST exhibited a U-shaped association with mortality risk. The AUC of the prediction model adding liver biomarkers into traditional prediction factors was 0.718 (95% CI: 0.679-0.757), with an increase of 0.030 ( P<0.001) compared with the traditional model. Conclusions:Severe hepatic steatosis, higher levels of hepatic fibrosis, and elevated GGT levels are significantly associated with higher mortality risk. AST shows a U-shaped nonlinear association with mortality risk. Incorporating liver biomarkers into traditional risk prediction models enhance the ability to predict mortality.

Objective:To investigate the prospective associations between plasma metabolites and the risks of all-cause and cause-specific mortality among Chinese adults.Methods:This study analyzed plasma metabolomics data from 2 183 healthy adults in the China Kadoorie Biobank (CKB), measured using targeted mass spectrometry. Cox proportional hazards regression models were used to examine the associations between 630 metabolites and the risk of all-cause mortality. Cause-specific hazard regression models evaluated the associations between metabolites and cardiovascular disease (CVD) risks, cancer, and other-cause mortality. Stepwise regression was used to identify key metabolites independently associated with all-cause mortality, and the area under the receiver operating characteristic curve (AUC) was calculated to assess the improvement in predictive performance when these metabolites were added to traditional risk prediction models.Results:The mean age of the participants was (53.2±9.8) years, 65.1% of whom were female. During a median follow-up of 14.5 years, 231 deaths occurred. A total of 44 metabolites were significantly associated with the risk of all-cause mortality [false discovery rate (FDR)-adjusted P<0.05], primarily including triglycerides, ceramides, and amino acids. Additionally, 29 and 15 metabolites were found to be associated with cancer and other-cause mortality, respectively, but no metabolites were significantly associated with CVD mortality after FDR corrections. Adding 14 metabolites independently associated with all-cause mortality into the traditional prediction model significantly improved its predictive performance. Specifically, incorporating metabolites into the traditional model, which already included laboratory biomarkers, increased the AUC to 0.798 (95% CI: 0.755-0.843), an improvement of 0.088 compared to the traditional model ( P<0.001). Conclusions:Multiple metabolites are significantly associated with mortality risk and can substantially improve the accuracy of mortality risk prediction models. These findings provide new insights into the physiological mechanisms of aging and offer valuable clues for personalized health risk assessment.

Objective:To explore the prospective association between physical activity level and mor-tality risk in Chinese adults with chronic obstructive pulmonary disease(COPD).Methods:Based on the China Kadoorie Biobank(CKB)who had COPD at the baseline survey,this study employed the Cox proportional hazards regression model to estimate the prospective associations between the overall physical activity,different intensities(low-level,moderate-to-vigorous-level),and types(occupational,non-occupational)of physical activity level and the risks of all-cause and cause-specific mortality,such as vascular diseases,cancer,and respiratory diseases.Based on the quintiles of physical activity level,par-ticipants were divided into five groups(Q1-Q5),with the lowest quintile group(Q1)as the reference group.Hazard ratio(HR)and 95％confidence interval(95％CI)were calculated for the remaining.In our study,we also performed sensitivity and subgroup analyses,including age,gender,self-rated health status,severity of COPD,etc.Results:Among 33 588 COPD patients at the baseline survey,8 314(22.2％)deaths were documented during an average follow-up of(11.1±3.1)years.Negative linear associations between the overall physical activity level and mortality risk from all-cause,vascular,and respiratory diseases were observed(P trend for linear correlation being＜0.001,0.002,＜0.001).Compared with the lowest quintile group of total physical activity(Q1),the hazard ratios(HR)and 95％confidence intervals(CI)for all-cause mortality,vascular disease mortality,and respiratory disease mortality in the highest quintile group(Q5)were 0.77(0.70,0.85),0.77(0.65,0.91),and 0.58(0.48,0.71),respectively.The low-level and moderate-to-vigorous-level physical activity were nega-tively associated with all-cause mortality in the COPD patients(P trend for linear correlation:0.002,＜0.001,respectively).Compared with the lowest quintile group of low-intensity and moderate-to-vigorous intensity physical activity(Q1),the HRs(95％CI)for all-cause mortality in the highest quintile group(Q5)were 0.89(0.82,0.97)and 0.79(0.72,0.87),respectively.The occupational and non-occupational physical activity were also found to have a linear inverse association with all-cause mortality risk among the COPD patients(P trend＜0.001 and 0.015,respectively).Compared with the lowest quintile group of occupational and non-occupational physical activity(Q1),the HR(95％CI)for all-cause mortality in the highest quintile group(Q5)were 0.69(0.61,0.78)and 0.91(0.84,0.98),respectively.The associations between overall physical activity and all-cause mortality risk were stronger for patients aged 60 and above,female,and who reported poor health status(P for interaction:0.028,0.012,0.010).The protective effect of total physical activity was also applicable to the COPD patients of varying severity.Conclusion:Physical activity could reduce the mortality risk in a dose-response relationship among COPD patients,regardless of its intensity and type,especially among indi-viduals aged 60 and above,females,and those with poor self-report health status.

Objective:To explore the impacts of smoking on the need for respiratory support and mortality in patients hospitalized with pneumonia.Methods:A total of 24 367 patients hospitalized with pneumonia from 2009 to 2017 in the China Kadoorie Biobank, were included. Smoking status was self-reported, and data regarding respiratory support during hospitalization and mortality during follow-up were obtained from medical claims and death registries, respectively. OR, HR, and 95% CI were calculated and adjusted for potential confounders using logistic regression models and Cox proportional hazards regression models, respectively. Results:Among males, current smokers or those who quit smoking due to illness had higher risks of requiring respiratory support ( OR=1.15, 95% CI: 1.03-1.29), 1-year mortality ( HR=1.66, 95% CI: 1.32-2.08), and 5-year mortality ( HR=1.32, 95% CI: 1.13-1.54) following pneumonia hospitalization compared to nonsmokers. Male smokers who started smoking at a younger age or with longer smoking duration had the highest mortality risks (trend test both P<0.05). Female current smokers or those who quit smoking due to illness had higher risks of 1-year mortality ( HR=1.62, 95% CI: 1.17-2.23) and 5-year mortality ( HR=1.33, 95% CI: 1.06-1.67). We found no statistically significant difference in 90-day mortality between current smokers/those who quit smoking due to illness and nonsmokers. Conclusions:Smoking was associated with higher risks of requiring respiratory support and mortality in patients hospitalized with pneumonia, especially among males and heavy smokers. These findings highlight the need for targeted strategies to promote smoking cessation in patients hospitalized with pneumonia.

Objective:To identify individuals with accelerated aging under the frailty index (FI) as a proxy indicator of biological age, and to investigate the associations of age at menarche, age at menopause, and reproductive lifespan with frailty status and multi-timepoint FI trajectories among Chinese adult women.Methods:The current study included 302 471 women from the China Kadoorie Biobank 2004-2008 baseline survey data. Their age at menarche and menopause were self-reported, and the duration of reproductive lifespan was calculated by subtracting the two ages. The baseline FI was constructed using 28 baseline variables, including diseases, symptoms, and anthropometric measurements. Frailty status was categorized into three groups: non-frail (FI≤0.10), pre-frail (0.10

Objective:To investigate the relationship between individual and combined exposure to ambient air pollutants and vitamin D, as well as the potential role of ultraviolet radiation (UVR) in mediating the relationships.Methods:This study included 6 967 participants from the China Kadoorie Biobank, whose baseline (2004-2008) blood 25-hydroxyvitamin D had been tested. The average monthly concentrations of pollutants (PM 2.5, PM 10, NO 2, O 3) and the UVR estimates in the month of blood sample collection were derived from the ChinaHighAirPollutants and ultraviolet datasets covering China, respectively. Linear models and multinomial logistic regression models were used to analyze the associations of each air pollutant with vitamin D concentrations and groups. Principal component analysis integrated with quantile-based g-computation was applied to evaluate the co-effects and relative contribution weights of air pollutants. Mediation analyses were performed to investigate the potential role of UVR. Results:PM 2.5, PM 10, NO 2, and O 3 were individually associated with vitamin D concentrations, as well as vitamin D groups: insufficiency (20-30 ng/ml) and deficiency (<20 ng/ml). For each 10 μg/m3 increase in monthly PM 2.5, PM 10, NO 2, and O 3, the percentage changes (95% CIs) in vitamin D concentrations were -1.71% (-2.16% - -1.26%), -1.30% (-1.60% - -1.00%), -3.77% (-4.60% - -2.93%), and 1.27% (0.91%-1.63%), respectively, with corresponding ORs (95% CIs) for vitamin D deficiency of 1.17 (1.06-1.29), 1.12 (1.05-1.19), 1.66 (1.38-2.00), and 0.81 (0.74-0.89). The mixture of the first principal component of PM 2.5 and PM 10 (PM), along with NO 2 and O 3, was negatively correlated with vitamin D. The percentage change (95% CI) in vitamin D concentrations for a one-quintile increase in the mixture was -2.20% (-3.56%- -0.82%), with NO 2 contributing the most (83%), followed by PM (17%). UVR-mediated association was 12.2% (95% CI：2.5%-23.0%) for PM 2.5 and 4.9% (95% CI：2.2%-8.0%) for PM 10 with vitamin D concentrations, respectively. Conclusion:Higher concentrations of particulate matter, NO 2, and a mixture of air pollutants were associated with lower vitamin D concentrations in Chinese adults, with reduced UVR acting as a partial mediator in the particulate matter-vitamin D relationship.

Objective:To investigate the relationship of several adiposity-related anthropometric parameters, including BMI, waist circumference (WC), waist-to-hip ratio (WHR), body fat percentage (BFP) and indoles in plasma with the incidence of atherosclerotic cardiovascular disease (ASCVD) in adults in China.Methods:In China Kadoorie Biobank (CKB) study, blood samples were collected from 2 183 participants in the first resurvey in 2008 to detect indoles. Participants' body weight, body height, WC, hip circumference, and BFP were measured at baseline survey in 2004 and resurvey in 2008, the BMI and WHR were calculated with standardized methods. The long-term follow-up of all participants started from the completion of the resurvey in 2008 until the occurrence of incident ASCVD, death, loss to follow-up or until December 31, 2018. CKB ascertained outcome status (incident ASCVD) through death and disease registries and national health insurance databases, supplemented by active follow-up. Multivariate linear regression model was used to estimate the associations of anthropometric measurements at baseline survey and the first resurvey, and changes in these measurements with 3 indoles [indole, indole-3-acetic acid (IAA), and indole-3-propionic acid (IPA)]. Cox proportional hazard regression model was used to estimate the associations between indoles and the risk for ASCVD.Results:Anthropometric measurements at baseline survey or the first resurvey were negatively associated with plasma IPA level. The regression coefficient ( β) of baseline BMI (per 1.0 kg/m 2) with 0.1 standard deviation ( SD) IPA was -0.23 (95% CI: -0.36 - -0.10) (false discovery rate=0.004). After adjusting for baseline BMI, the β of baseline WC, WHR and BFP with 0.1 SD IPA were -0.09 (95% CI: -0.18 - -0.01), -0.12 (95% CI: -0.19 - -0.05), and -0.20 (95% CI: -0.32 - -0.08), respectively. The annual change in BMI (difference between BMI in 2008 and 2004 divided by the time interval) was associated with indole and IAA, with β of 1.40 (95% CI: 0.58 - 2.21) and -1.07 (95% CI: -1.91 - -0.23), respectively, at each 0.1 increase of SD. Over a median ( Q1, Q3) follow-up of 10.46 (10.36, 10.53) years after 2008 resurvey, 236 cases of ASCVD were recorded. IAA and IPA levels were negatively associated with the risk for ASCVD, with hazard ratios for one SD increase of IAA and IPA of 0.87 (95% CI: 0.76 - 0.99) and 0.84 (95% CI: 0.73 - 0.96), respectively. Conclusions:Our results suggested that anthropometric measurements and their changing trends affect the levels of plasma imicrobial tryptophan metabolite levels, decreased levels of IAA and IPA levels are associated with increased risk of ASCVD and indoles in plasma including IPA and IAA might be the mediating factors for adiposity-induced ASCVD.

Objective:To analyze the prospective associations between liver biomarkers and mortality among Chinese middle-aged and elderly populations and to evaluate the mortality risk predictive value.Methods:A total of 22 758 participants from the 3 rd resurvey of the China Kadoorie Biobank were included. Cox proportional hazard models were used to analyze the prospective associations of 5 liver biomarkers with mortality. These liver biomarkers included two liver imaging biomarkers (liver fat attenuation parameter, liver stiffness measurement) and three serum liver enzyme biomarkers [gamma-glutamyl transferase (GGT), ALT, and AST]. Restricted cubic spline was used to assess the nonlinear associations between biomarkers and mortality. The area used the receiver operating characteristic curve (AUC) to evaluate the predictive ability of the models after incorporating liver biomarkers into traditional prediction models for mortality. Results:The mean age of the participants was (65.2±9.1) years, with a median follow-up of 1.5 years, during which 307 deaths occurred. Compared to individuals without hepatic steatosis, those with severe hepatic steatosis had a 79% higher risk of mortality, with a HR of 1.79 (95% CI: 1.06-3.03). Compared to individuals without hepatic fibrosis, those with advanced fibrosis and cirrhosis had higher mortality risks of 48% and 91%, respectively (both P<0.05). For each standard deviation increase in GGT, the mortality risk increased by 10% ( HR=1.10, 95% CI: 1.05-1.15), with the positive association plateauing at higher GGT levels. AST exhibited a U-shaped association with mortality risk. The AUC of the prediction model adding liver biomarkers into traditional prediction factors was 0.718 (95% CI: 0.679-0.757), with an increase of 0.030 ( P<0.001) compared with the traditional model. Conclusions:Severe hepatic steatosis, higher levels of hepatic fibrosis, and elevated GGT levels are significantly associated with higher mortality risk. AST shows a U-shaped nonlinear association with mortality risk. Incorporating liver biomarkers into traditional risk prediction models enhance the ability to predict mortality.

Objective:To investigate the prospective associations between plasma metabolites and the risks of all-cause and cause-specific mortality among Chinese adults.Methods:This study analyzed plasma metabolomics data from 2 183 healthy adults in the China Kadoorie Biobank (CKB), measured using targeted mass spectrometry. Cox proportional hazards regression models were used to examine the associations between 630 metabolites and the risk of all-cause mortality. Cause-specific hazard regression models evaluated the associations between metabolites and cardiovascular disease (CVD) risks, cancer, and other-cause mortality. Stepwise regression was used to identify key metabolites independently associated with all-cause mortality, and the area under the receiver operating characteristic curve (AUC) was calculated to assess the improvement in predictive performance when these metabolites were added to traditional risk prediction models.Results:The mean age of the participants was (53.2±9.8) years, 65.1% of whom were female. During a median follow-up of 14.5 years, 231 deaths occurred. A total of 44 metabolites were significantly associated with the risk of all-cause mortality [false discovery rate (FDR)-adjusted P<0.05], primarily including triglycerides, ceramides, and amino acids. Additionally, 29 and 15 metabolites were found to be associated with cancer and other-cause mortality, respectively, but no metabolites were significantly associated with CVD mortality after FDR corrections. Adding 14 metabolites independently associated with all-cause mortality into the traditional prediction model significantly improved its predictive performance. Specifically, incorporating metabolites into the traditional model, which already included laboratory biomarkers, increased the AUC to 0.798 (95% CI: 0.755-0.843), an improvement of 0.088 compared to the traditional model ( P<0.001). Conclusions:Multiple metabolites are significantly associated with mortality risk and can substantially improve the accuracy of mortality risk prediction models. These findings provide new insights into the physiological mechanisms of aging and offer valuable clues for personalized health risk assessment.