標題: 台灣地區細懸浮微粒(PM2.5)對健康的影響研究
Association of fine particles (PM2.5) with health effect in Taiwan
作者: 李育慈
Lee, Yu-Cih
蔡春進
工學院永續環境科技學程
關鍵字: 死亡率;平均餘命;多變數迴歸分析;PM2.5
公開日期: 2013
摘要: 國內外已有許多PM2.5(細懸浮微粒)的健康影響研究,然而國內的研究受限於PM2.5的監測數據尚未普及,取得數據的地區較少且時間較短的因素,因此有必要再進一步蒐集更多的PM2.5質量濃度及各疾病死亡率的數據,以進行更完整的台灣地區PM2.5的健康影響研究。本研究蒐集94年至100年台灣本島各縣市的PM2.5質量濃度、家庭收入、人口資料、吸煙盛行率、平均餘命及疾病死亡率等6種資料,並分成全部年齡層、25歲以下年齡層與25歲至65歲年齡層等3個,及全死因死亡率、全癌症死亡率、心血管及呼吸道相關疾病的死亡率以及3種單一疾病(高血壓性疾病、腦血管疾病、肺炎)的死亡率等4種死亡率類別後,以西部19個縣市的數據資料來進行4個模式(Model 1~4)的多變數線性迴歸分析,以了解PM2.5對各年齡族群平均餘命及各疾病的死亡率所產生的影響。 死亡率的多變數迴歸分析結果顯示(Model 1),只有全死因死亡率與25歲至65歲年齡層的3個自變數(PM2.5年平均濃度、家庭收入和吸煙盛行率)具相關性。相對之下,在平均餘命的分析結果顯示(Model 2),平均餘命與3個自變數(PM2.5年平均濃度、家庭收入和吸煙盛行率),在3個年齡層及死亡率類別全部皆呈相關性且相關性高(R2=0.76~0.90)。在相對風險的影響上,死亡率的相對風險分析結果顯示(Model 3),死亡率的相對風險不會與任何一個或一組自變數呈現相關性。平均餘命的相對風險分析結果顯示(Model 4),平均餘命的相對風險仍然不會與任何自變數呈現相關性。 歸納以上結果,PM2.5對平均餘命(Model 2)比死亡率(Model 1)較具相關性,但對兩者之相對風險(Model 3、4)的相關性較低。 關鍵詞: PM2.5,死亡率,平均餘命,多變數迴歸分析
There are many researches on the health effects of PM2.5 (fine particles) in the nation and countries around the world. However, the domestic research is limited to the inadequate period of PM2.5 monitoring data, a few districts and shorter period, therefore there is a need to acquire more PM2.5 monitoring and mortality data to provide a more comprehensive analysis of the PM2.5 health effect in Taiwan. This study collected PM2.5 monitoring, family income, population, fraction of tabacco smoking, life expectancy and mortality data in all counties and cities in Taiwan. These six kinds of data were divided into three age groups and four motality kinds to conduct multivrariate regression using four models (model 1~4) in 19 counties and cities in Western Taiwan. The research helps to understand the influence of PM2.5 on the effect of life expectancy and mortality due to diseases. In the results of the multivariate analysis of mortality (Model 1), only the all-cause mortality is relevant to three independent variables (annual PM2.5 concentration, family income and fraction of tobacco smoking) at the 25 to 65 age group. In comparison, in the life expectancy multivariate analysis (Model 2), life expectancy and independent variables (annual PM2.5 concentration, family income, and fraction of tobacco smoking) in the three age groups and all four motality kinds are highly relevant (R2=0.76~0.90). In the relative risk of mortality multivariate analysis (Model 3), none of the independent variables in the three independent variables exhibit the relevance. In the relative risk of life expectancy multivariate analysis (Model 4), there are still no independent variables which are relevant. To sum up, PM2.5 has more influence on the life expectancy (model 2) than the mortality (model 1), but is less relevant to the relative risk of both life expectancy and mortality (Model 3 and 4). Keywords: PM2.5, mortality, life expectancy, multi-variable regression analysis
URI: http://140.113.39.130/cdrfb3/record/nctu/#GT070061516
http://hdl.handle.net/11536/73678
Appears in Collections:Thesis