Statistical methods for linking the chemical composition of particulate matter to health outcomes

Air pollution;health effects;particulate matter;statistics;not listed

Krall, Jenna RoseBreysse, Patrick N. ;

Johns Hopkins University

关键词: Air pollution;    health effects;    particulate matter;    statistics;    not listed;   

Others  :  https://jscholarship.library.jhu.edu/bitstream/handle/1774.2/37031/KRALL-DISSERTATION-2014.pdf?sequence=1&isAllowed=y

瑞士|英语

来源: JOHNS HOPKINS DSpace Repository

【 摘 要 】

Short-term exposure to particulate matter air pollution less than 2.5 micrometers in aerodynamic diameter (PM2.5) has been associated with mortality and morbidity in epidemiologic studies.PM2.5 is a complex mixture of many different chemicals that are emitted from both natural and anthropogenic sources.Recent studies have found that the toxicity of PM2.5 depends on its chemical composition, however estimating ambient PM2.5 constituent concentrations is challenging in part for several reasons.First, the network of ambient monitors that measure PM2.5 chemical constituents is spatially sparse and many communities have only one monitor.Additionally, some constituents of PM2.5 are spatially heterogeneous and their concentrations observed at ambient monitors may not be representative of surrounding areas.Last, PM2.5 constituents that contribute minimally to PM2.5 total mass frequently have censored concentrations that fall below minimum detection limits.Together, these attributes of the available data make estimating ambient PM2.5 constituent concentrations difficult and complicate subsequent health effects analyses.We could also estimate health effects of PM2.5 sources, which emit combinations of chemical constituents. PM2.5 sources are not directly measured and are frequently inferred from PM2.5 constituent concentrations. The challenge of estimating PM2.5 sources is magnified by the measurement error and data limitations of observed PM2.5 constituent concentrations. This dissertation developed methods to address measurement issues in estimating ambient concentrations of PM2.5 constituents and PM2.5 sources.Then, these methods were used to estimate associations between mortality and short-term exposure to PM2.5 constituents and PM2.5 sources.