Estimated hourly personal exposures to ambient and nonambient particulate matter among sensitive populations in Seattle, Washington
Estimated hourly personal exposures to ambient and nonambient particulate matter among sensitive populations in Seattle, Washington
复制标题
DOI：
10.1080/10473289.2004.10470988
复制
发表时间：
2004-09-01
影响因子：
2.7
通讯作者：
Liu, LJS
中科院分区：
文献类型：
--
作者：
Allen, R;Wallace, L;Liu, LJS
颗粒物（PM）的流行病学研究通常使用固定式室外监测仪测量的浓度作为个人暴露的替代品。尽管经常报道环境浓度与个人暴露总量之间的相关性较差，但环境浓度与健康影响之间的流行病学关联取决于环境浓度与个人暴露于环境产生的PM之间的相关性。本文将个人PM暴露分为环境和非环境成分，并利用收集的38名受试者的连续光散射数据估计室外对个人PM暴露的贡献。西雅图,华盛顿州。在所有的研究对象中，在室内遇到的平均暴露量低于所有其他微环境。做饭和上学都与高水平的接触有关。先前公布的粒子浸润（F-inf）估计值与时间-位置数据相结合，以估计每个受试者的环境贡献分数（alpha，平均值= 0.66 +/- 0.21）。采暖季监测对象的平均a值（0.55 +/- 0.16）显著低于非采暖季监测对象（0.80 +/- 0.17）。我们的模型alpha估计值与硫示踪剂方法估计值很好地吻合（斜率= 1.08;R-2 = 0.67）。我们用连续光散射和24小时重力数据模拟了暴露于环境和非环境PM的情况，发现两种方法之间有很好的一致性。平均而言，环境颗粒占个人总暴露量的48%（范围= 21-80%）。个人活动暴露受到远离监测微环境的时间的高度影响。中心站点浓度与个人暴露之间的每小时纵向相关性中位数为0.30。虽然环境源和非环境源都影响个人-中心关系，但后者似乎占主导地位。因此，固定的室外监测仪可能很难预测个人暴露总量，特别是在PM暴露主要由非环境暴露的人群中，例如，居住在密闭房屋中的人、做饭的人和儿童。
Epidemiological studies of particulate matter (PM) routinely use concentrations measured with stationary outdoor monitors as surrogates for personal exposure. Despite the frequently reported poor correlations between ambient concentrations and total personal exposure, the epidemiologic associations between ambient concentrations and health effects depend on the correlation between ambient concentrations and personal exposure to ambient-generated PM. This paper separates personal PM exposure into ambient and nonambient components and estimates the outdoor contribution to personal PM exposures with continuous light scattering data collected from 38 subjects in. Seattle, WA. Across all subjects, the average exposure encountered indoors at home was lower than in all other microenvironments. Cooking and being at school were associated with elevated levels of exposure. Previously published estimates of particle infiltration (F-inf) were combined with time-location data to estimate an ambient contribution fraction (alpha, mean = 0.66 +/- 0.21) for each subject. The mean a was significantly lower for subjects monitored during the heating season (0.55 +/- 0.16) than for those monitored during the nonheating season (0.80 +/- 0.17). Our modeled alpha estimates agreed well with those estimated with the sulfur-tracer method (slope = 1.08; R-2 = 0.67). We modeled exposure to ambient and nonambient PM with both continuous light scattering and 24-hr gravimetric data and found good agreement between the two methods. On average, ambient particles accounted for 48% of total personal exposure (range = 21-80%). The personal activity exposure was highly influenced by time spent away from monitored microenvironments. The median hourly longitudinal correlation between central site concentrations and personal exposures was 0.30. Although both a and the nonambient sources influence the personal-central relationship, the latter seems to dominate. Thus, total personal exposure may be poorly predicted by stationary outdoor monitors, particularly among persons whose PM exposure is dominated by nonambient exposures, for example, those living in tightly sealed homes, those who cook, and children.