Abstract
BACKGROUND: Many studies have explored the association between Body Mass Index (BMI) trajectory and mortality, but the association between BMI trajectory during old age and mortality remains underreported, particularly in the Chinese population. This study aimed to investigate the association between BMI trajectories in older adulthood and all-cause mortality in China, and to analyze potential mediating mechanisms. METHODS: We analyzed data from the Chinese Longitudinal Healthy Longevity Survey with latent class growth modeling to identity BMI trajectories at 3 follow-up visits (2008, 2011 and 2014). All-cause mortality was assessed from baseline to July 31,2019. Cox proportional hazard regression was used to estimate the association between BMI trajectories and all-cause mortality. RESULTS: Among 3676 older adults (female: 52.3%, median (IQR) age was 77 (70, 85) years), after 12,516 person-years of follow-up, 1,331 all-cause deaths were recorded. Three distinct BMI trajectories were identified: low-normal stable trajectory (47.33%), normal slight increase trajectory (44.45%), and overweight to obesity trajectory (8.22%). In the fully adjusted model, compared to the normal slight increase trajectory, the risk of all-cause mortality was significantly increased in the low-normal stable trajectory (HR = 1.39, 95%CI: 1.22, 1.60), while the risk of mortality was not statistically different in the overweight to obesity trajectory (HR = 1.16, 95%CI: 0.83, 1.61). Both stratified and sensitivity analyses confirmed these findings. Mediation analysis suggested that cognitive impairment and lack of leisure activities might partially mediate this association. Threshold analysis indicated that the risk of mortality gradually decreases with increasing BMI when BMI is below 26 kg/m(2) (HR = 0.95, 95%CI: 0.93, 0.97) and then remains stable after 26 kg/m(2). CONCLUSION AND RELEVANCE: Compared with normal slight increase trajectory, low-normal stable BMI trajectory during old age may increase the risk of all-cause mortality. These insights hold significant implications for future health management strategies and interventions, aiming to enhance the overall health status and quality of life among older adults.