Abstract
BACKGROUND: Based on the previous evidence, physical function has been associated with all-cause mortality. However, these studies have been inconsistent. We aimed to conduct trajectory analysis to identify instrumental activities of daily living (IADL) types and estimate their effects on all-cause mortality among older people. METHODS: In the Chinese Longitudinal Healthy Longevity Survey, a total of 13 385 older people aged ≥60 years were included between 2002-18. We employed a group-based trajectory model to determine the IADL trajectories. We fitted a multivariate Cox regression model to evaluate the effects of various IADL trajectories on all-cause mortality. We applied subgroup analyses to explore potential modified effects. We further conducted sensitivity analyses to ascertain the robustness of findings. RESULTS: Over the 16-year follow-up period, three IADL trajectories were identified, including 'stable and high function' (45.7%), 'rapid increase' (25.5%), and 'stable and low function' (28.7%). The Cox regression model manifested that 'stable and low function' was positively associated with high risk of all-cause mortality compared with 'stable and high function' (hazard ratio (HR) = 1.33; 95% confidence interval (CI) = 1.25-1.41). Subgroup analyses indicated that this association was also modified by age, income, marital status, social activity, and cognitive impairment (P < 0.05). Additionally, our findings remained robust after excluding individuals with chronic diseases or mild cognitive impairment at baseline (P < 0.05). CONCLUSIONS: Premature mortality among older people is associated with stable and low IADL function. Additionally, our findings suggested that public health policies should further focus on maintaining functional ability in relatively healthy older people.