Long-Term Ozone Exposure and Mortality in a Large Prospective Study

Abstract

Rationale: Tropospheric ozone (O 3 ) is potentially associated with cardiovascular disease risk and premature death. Results from longterm epidemiological studies on O 3 are scarce and inconclusive.Objectives: In this study, we examined associations between chronic ambient O 3 exposure and all-cause and cause-specific mortality in a large cohort of U.S. adults.Methods: Cancer Prevention Study II participants were enrolled in 1982. A total of 669,046 participants were analyzed, among whom 237,201 deaths occurred through 2004. We obtained estimates of O 3 concentrations at the participant's residence from a hierarchical Bayesian space-time model. Estimates of fine particulate matter (particulate matter with an aerodynamic diameter of up to 2.5 mm [PM 2.5 ]) and NO 2 concentrations were obtained from land use regression. Cox proportional hazards regression models were used to examine mortality associations adjusted for individual-and ecological-level covariates. Measurements and Main Results:In single-pollutant models, we observed significant positive associations between O 3 , PM 2.5 , and NO 2 concentrations and all-cause and cause-specific mortality. In two-pollutant models adjusted for PM 2.5 , significant positive associations remained between O 3 and all-cause (hazard ratio [HR] per 10 ppb, 1.02; 95% confidence interval [CI], 1.01-1.04), circulatory (HR, 1.03; 95% CI, 1.01-1.05), and respiratory mortality (HR, 1.12; 95% CI, 1.08-1.16) that were unchanged with further adjustment for NO 2 . We also observed positive mortality associations with both PM 2.5 (both near source and regional) and NO 2 in multipollutant models.Conclusions: Findings derived from this large-scale prospective study suggest that long-term ambient O 3 contributes to risk of respiratory and circulatory mortality. Substantial health and environmental benefits may be achieved by implementing further measures aimed at controlling O 3 concentrations.

The influence of model spatial resolution on simulated ozone and fine particulate matter for Europe: implications for health impact assessments

“…Since, seasonal differences in simulated O 3 with resolution are considerable, the AF associated with long-term exposure to O 3 was also calculated based on annual-mean (as opposed to summer-mean) O 3 concentrations, which were in turn based on recommendations from Turner et al (2015). Turner et al (2015) suggest a higher concentration response coefficient of 1.06 (95 % CI: 1.04-1.08) per 10 µg m −3 and a slight lower MDA8 O 3 threshold of 53.4 µg m −3 compared to values used in our study for summer-mean MDA8 O 3 .…”

“…A substantial number of epidemiological studies have derived risk estimates for mortality associated with long-term exposure to ambient fine particulate matter with an aerodynamic diameter of less than 2.5 µm (PM 2.5 ) Brook et al, 2010;WHO, 2013) and also recently, to a lesser extent, for long-term exposure to ozone (O 3 ) Forouzanfar et al, 2016;Turner et al, 2015). Differences in risk estimates produced from different epidemiological studies can be due to differences in methodologies, air pollution and health data used, including the size and spatial extent of cohort populations.…”

“…Markakis et al, 2015;Schaap et al, 2015;Yu et al, 2016;Neal et al, 2017). Although the same model processes are represented at different model resolutions, simulated pollutant concentrations can vary due to differences in (i) the resolution of emissions, which may have a nonlinear effect on the chemical formation of pollutants, and (ii) the resolution of the driving meteorology (Valari and Menut, 2008;Tie et al, 2010;Arunachalam et al, 2011;Colette et al, 2013;Markakis et al, 2015;Schaap et al, 2015).…”

“…The impact of model horizontal resolution on simulated O 3 concentrations has been primarily linked to less dilution of emissions when using a finer resolution (Valari and Menut, 2008;Tie et al, 2010;Colette et al, 2013;Stock et al, 2014;Schaap et al, 2015). Investigating the impact of increasing model horizontal resolution from 48 to 6 km on O 3 concentrations in Paris, Valari and Menut (2008) found modelled surface O 3 to be more sensitive to the resolution of input emissions than to meteorology.…”

“…Investigating the impact of increasing model horizontal resolution from 48 to 6 km on O 3 concentrations in Paris, Valari and Menut (2008) found modelled surface O 3 to be more sensitive to the resolution of input emissions than to meteorology. A number of other studies note the sensitivity of simulated O 3 to simulated nitrogen oxide (NO x ) concentrations that determine the extent of titration of O 3 by nitrogen monoxide (NO) (Stock et al, 2014;Markakis et al, 2015;Schaap et al, 2015).…”