Associations between Air Pollution and Daily Emergency Room Visits for Cardio-Respiratory Diseases

許多研究指出暴露於微粒空氣污染物對死亡率有增加的情形，尤其對心臟血管疾病及肺部疾病的影響更是明顯，也有研究指出空氣污染物的暴露亦會增加急診求診率，但是極少研究探討微粒數量濃度以及微粒成分對健康的影響。本研究的目的為探討微粒數量濃度以及微粒成分與急診求診之相關。研究結合2004年1月1日至2008年8月31日國立臺灣大學醫學院附設醫院心肺疾病急診求診資料與空氣品質監測站資料（包括超級測站及一般空品測站），利用年份、月份及星期等時間變項，以及溫度、濕度、雨量等氣候變項，建立不同疾病不包含空氣污染物的基本模型(base model)，再利用廣義加法模型(generalized additive models, GAM）評估單一空氣污染物(single-pollutant)與多空氣污染物(multi-pollutant)從當日到前五日(lag 0-5 days)以及從當天與前一天的空氣污染物濃度平均值到當天與前四天的空氣污染物濃度平均值(2-day to 5-day moving average)對於心血管(國際疾病傷害及死因分類標準第九版ICD-9 codes: 390-459以及785)及呼吸道疾病(國際疾病傷害及死因分類標準第九版ICD-9 codes: 460-519以及786)急診求診情形的影響。研究結果是以每增加一個四分位數間距(interquartile range, IQR)的空氣污染物會增加多少急診求診的相對危險性來表示。分析所使用到的空氣污染物成分包括：PM10、PM2.5、硝酸鹽 (nitrate)、硫酸鹽 (sulfate)、黑碳 (black carbon)、有機碳 (organic carbon)以及元素碳 (element carbon)。在單一污染物模型(single-pollutant model)中顯示，微粒成分對於心血管疾病、呼吸道疾病的急診情形有顯著的影響(P < 0.05)。進一步利用多空氣污染物模型(multi-pollutant model) 校正一氧化碳(CO)與臭氧(O3)等氣狀污染物與PM10或PM2.5後發現，硝酸鹽、硫酸鹽、黑碳、有機碳以及元素碳與呼吸道急診有關，其最大相對危險性分別是1.025、1.025、1.023、1.031以及1.012。在心血管疾病方面則發現硫酸鹽、有機碳與元素碳與急診有關，最大相對危險性分別是1.020、1.061以及1.017。再將急診資料依據年齡分成孩童(0-14歲)、成年人(15-64歲)、老年人(> 65歲)等三組進行分析，發現硝酸鹽、硫酸鹽、黑碳、有機碳以及元素碳對成年人呼吸道疾病急診有顯著影響，最大相對危險性分別是1.031、1.043、1.037、1.050以及1.020。而硫酸鹽、有機碳與元素碳則對心血管疾病有影響，最大相對危險性為1.068、1.060以及1.036；在孩童方面，也發現硝酸鹽、硫酸鹽、黑碳、有機碳以及元素碳對呼吸道疾病急診的影響，最大相對危險性分別是1.036、1.020、1.030、1.046以及1.015；而在老年人只看到硝酸鹽與有機碳當天的濃度(lag 0 day)會增加心血管疾病急診的危險性，最大相對危險性為1.017以及1.050。硫酸鹽與有機碳則發現與老年人呼吸道疾病急診有關，最大相對危險性是1.045以及1.032。本研究結果顯示微粒的成份與心血管及呼吸道疾病急診危險性的增加有關，而且對成年人的影響較大。Many studies indicate that urban air pollution may increase the risk of mortality and mobility for cardio-respiratory diseases, but most of these studies focus on particulate matter mass concentrations. Epidemiological studies examined the impact of particulate matter number concentrations and components on health effects were very few. The objective of this study is to examine the association between particle number concentrations and particulate components and emergency room visits for cardiovascular and respiratory diseases in Taipei metropolitan area. We linked the environmental data obtained from the Supersite in Sinjhuang and air monitoring stations in Taipei metropolitan area with emergency room visits for cardiovascular (ICD-9 codes: 390-459 and 785) and respiratory (ICD-9 codes: 460-519 and 786) diseases in National Taiwan University Hospital form 2004 January 1 to 2008 August 31. We first built base models for different diseases categories, including cardiovascular diseases and respiratory disease, without air pollution. The time-trend variables used in the base models included year, month and day-of-week. The meteorological variables included temperature, relative humidity and precipitation. Generalized additive models (GAM) were applied from lag 0 to 5 days and 2-day to 5-day moving average in our statistical analysis. Study results were expressed with relative risks (RRs) for an IQR (interquartile ranges) increase in each air pollutant. The particulate components analyzed in this study included nitrate, sulfate, black carbon (BC), organic carbon (OC) and element carbon (EC). Study results showed that particulate components were significantly associated with cardiovascular and respiratory diseases in single-pollutant models. We adjusted the effects of gaseous pollutants, carbon monoxide (CO) and ozone (O3), and PM10 or PM2.5 together to build multi-pollutant models. The associations between nitrate, sulfate, BC, OC and EC and respiratory diseases were still observed. The maximum RRs for each pollutant were 1.025, 1.025, 1.023, 1.031and 1.012, respectively. But for cardiovascular diseases, the associations were only found for sulfate, OC and EC with the maximum RRs were 1.020, 1.061 and 1.017. We further divided emergency room visits into three age groups, children (age 0-14 years), adults (age 15-64 years) and elderly (age above 65 years), to examine the effects of components. Nitrate, sulfate, BC, OC and EC were associated with emergency room visits for respiratory disease in adults group. The maximum RRs were 1.031, 1.043, 1.037, 1.050 and 1.020, respectively. For cardiovascular diseases in adults, the relations were only found in sulfate, OC and EC with the maximum RRs were 1.068, 1.060 and 1.036. In children, the associations between particles components and respiratory diseases were observed. The maximum RRs for nitrate, sulfate, BC, OC and EC were 1.036, 1.020, 1.030, 1.046 and 1.015. However, the impacts of particles components on cardiovascular diseases for elderly were only observed at lag 0 day for nitrate and OC. The maximum RRs were 1.017 and 1.050. Sulfate and OC were found increase the risk of respiratory diseases with the maximum RRs were 1.045 and 1.032. Those results indicate that particulate components have important influence on emergency room visits for cardiovascular and respiratory diseases, and the stronger effects were observed in adults group.第1章 前言 1.1 文獻回顧 1.1.1. 微粒重量濃度與健康效應的相關流行病學研究 1.1.2. 微粒數量濃度與健康效應的相關流行病學研究 3.1.3. 微粒成分與健康效應的相關流行病學研究 4.1.4. 氣狀空氣污染物與健康效應的相關流行病學研究 6.2. 研究目的 72章 材料與方法 8.1. 研究架構 8.2. 環境資料來源 9.3. 急診資料來源 11.4. 統計分析 133章 研究結果 17.1. 環境監測資料與急診資料描述 17.2. 基本模型的建立 26.2.1. 心血管疾病 26.2.2. 呼吸道疾病 27.3. 空氣污染物及急診求診之相關 29.3.1. 微粒數量濃度與微粒成分的單一污染物模型 29.3.2. 微粒成分的多污染物模型 34.3.3. 微粒成分對於不同年齡層的影響 394章 討論 575章 參考文獻 6