Excess deaths and loss of life expectancy attributed to long-term NO₂ exposure in the Chinese elderly

BACKGROUND: Evidence linking nitrogen dioxide (NO₂) air pollution to life span of high-vulnerability older adults is extensively scarce in low- and middle-income countries. This study seeks to quantify mortality risk, excess deaths, and loss of life expectancy (LLE) associated with long-term exposure to NO₂ among elderly individuals in China. METHODS: A nationwide dynamic cohort of 20352 respondents ≥65 years old were enrolled from the China Longitudinal Health and Longevity Survey during 2005-2018. Residential exposures to NO₂ and co-pollutants were assessed by well-validated spatiotemporal prediction models. A Cox regression model with time-dependent covariates was utilized to quantify the association of all-cause mortality with NO₂ exposure, controlling for confounders such as demographics, lifestyle, health status, and ambient temperature. NO₂-attributable deaths and LLE were evaluated for the years 2010 and 2020 based on the pooled NO₂-mortality relation derived from multi-national cohort investigations. Decomposition analyses were conducted to dissociate net shift in NO₂-related deaths between 2010 and 2020 into four primary contributing factors. RESULTS: A total of 14313 deaths were recorded during follow-up of approximately 100 hundred person-years (median 3.6 years). We observed an approximately linear relationship (nonlinear P = 0.882) of NO₂ exposure with all-cause death across a broad range from 6.6 to 95.7 μg/m³. Every 10-μg/m³ rise in yearly average NO₂ concentration was linked to a hazard ratio (HR) of 1.045 (95% confidence interval [CI]: 1.031-1.059). In the updated meta-analysis of this study and 9 existing cohorts, we estimated a pooled HR of 1.043 (95% CI: 1.023-1.063) for each 10-μg/m³ growth in NO₂. Reaching a 10-μg/m³ counterfactual target of NO₂ concentration in China could avoid 0.33 (95% empirical CI: 0.19-0.49) million premature deaths and an LLE of 0.40 (95% empirical CI: 0.23-0.59) years in 2010, which greatly dropped to 0.24 (95% empirical CI: 0.14-0.36) million deaths and 0.21 (95% empirical CI: 0.12-0.31) years of LLE in 2020. The net fall in NO₂-attributable deaths (-26.8%) between 2010 and 2020 was primarily driven by the declines in both NO₂ concentration (-41.6%) and mortality rate (-27.1%) under population growth (+41.0%) and age structure transition (+0.9%). CONCLUSIONS: Our findings provide national evidence for increased risk of premature death and loss of life expectancy attributed to later-life NO₂ exposure among the elderly in China. In an accelerated aging society, strengthened clean air actions should be formulated to minimize the health burden and regional inequality in NO₂-attributable mortality.