Mortality Risk and Burden From a Spectrum of Causes in Relation to Size-Fractionated Particulate Matters: Time Series Analysis

There is limited evidence regarding the adverse impact of particulate matters (PMs) on multiple body systems from both epidemiological and mechanistic studies. The association between size-fractionated PMs and mortality risk, as well as the burden of a whole spectrum of causes of death, remains poorly characterized. We aimed to examine the wide range of susceptible diseases affected by different sizes of PMs. We also assessed the association between PMs with an aerodynamic diameter less than 1 µm (PM ), 2.5 µm (PM ), and 10 µm (PM ) and deaths from 36 causes in Guangzhou, China. Daily data were obtained on cause-specific mortality, PMs, and meteorology from 2014 to 2016. A time-stratified case-crossover approach was applied to estimate the risk and burden of cause-specific mortality attributable to PMs after adjusting for potential confounding variables, such as long-term trend and seasonality, relative humidity, temperature, air pressure, and public holidays. Stratification analyses were further conducted to explore the potential modification effects of season and demographic characteristics (eg, gender and age). We also assessed the reduction in mortality achieved by meeting the new air quality guidelines set by the World Health Organization (WHO). Positive and monotonic associations were generally observed between PMs and mortality. For every 10 μg/m increase in 4-day moving average concentrations of PM , PM , and PM , the risk of all-cause mortality increased by 2.00% (95% CI 1.08%-2.92%), 1.54% (95% CI 0.93%-2.16%), and 1.38% (95% CI 0.95%-1.82%), respectively. Significant effects of size-fractionated PMs were observed for deaths attributed to nonaccidental causes, cardiovascular disease, respiratory disease, neoplasms, chronic rheumatic heart diseases, hypertensive diseases, cerebrovascular diseases, stroke, influenza, and pneumonia. If daily concentrations of PM , PM , and PM reached the WHO target levels of 10, 15, and 45 μg/m , 7921 (95% empirical CI [eCI] 4454-11,206), 8303 (95% eCI 5063-11,248), and 8326 (95% eCI 5980-10690) deaths could be prevented, respectively. The effect estimates of PMs were relatively higher during hot months, among female individuals, and among those aged 85 years and older, although the differences between subgroups were not statistically significant. We observed positive and monotonical exposure-response curves between PMs and deaths from several diseases. The effect of PM was stronger on mortality than that of PM and