Effects of long-term exposure to metal components of particulate matter and mortality in a large cohort

Abstract

Introduction: We aimed to study the association between long-term exposure to PM10, PM2.5 and several constituents of particulate matter, and non-accidental, cardiovascular (CVD) and ischemic heart disease (IHD) mortality. Methods: Using alarge administrative cohort study, we selected individuals aged 30 and olderat inclusion (October 2001) and followed them until December 2010. We used land use regression models to estimate annual average concentrations of air pollution at residence. We considered copper, iron, and zinc as tracers of non-tailpipe traffic emission pollution, sulphur for long-range transport, silicon for crustal material, potassium for biomass burning, and nickel and vanadium for mixed oil burning and industry. We used Cox models to estimate the associations between pollutants and cause-specific mortality, adjusting for individual and contextual characteristics. We estimated hazard ratios for 5th-95th percentile range increase in pollutants. Results: A total of 1,249,108 residents (54.5% female) were included in the study. There was evidence of association between all pollutants and non-accidental, CVD, and IHD mortality. With increasing PM2.5 we found a 3%, 5% and 6% higher risk in non-accidental, CVD and IHD mortality, respectively. We found high mortality risk with increased non-tailpipe traffic emission tracers: for PM2.5zinc we found a 6% (95%CI: 4-8%), 7% (95%CI: 4-10%) and 11% (95%CI: 6-16%) increased risk in non-accidental mortality, CVD, and IHD mortality, respectively. Exposure to mixed oil burning/industry tracers was also harmful to health. With increased nickel in PM10, we found a 7%, 8% and 13% higher risk in non-accidental, CVD and IHD mortality, respectively. Results were robust when we adjusted for PM mass. Conclusions: This study indicates an important role of PM metals originating from resuspension of road dust and mixed oil burning/industry tracers.