透過您的圖書館登入
IP:18.221.85.33
  • 學位論文

台灣北部地區肺炎及流行性感冒之盛行與風險分析

Prevalence and Risk Analysis for Outpatients of Pneumonia and Influenza in Northern Taiwan

指導教授 : 王玉純

摘要


本研究目的為探討北部地區(台北縣市、桃園縣)肺炎及流行性感冒(Pneumonia & Influenza, P&I)之盛行率與風險因子分析。研究資料包括全民健康保險資料庫1996-2007年每日就醫資料(ICD-9 CM 480-486, 487)、疾病管制局合約實驗室1999-2008年 類流感病毒檢測結果、1994-2007年八項指標性空氣污染物(CO、NO2、NO、NOX、PM10、SO2、O3, 8h、O3, 24h)每日濃度資料和1996-2007年溫度、相對濕度等每日氣象資料。 我們運用貝氏二階層模型(Bayesian 2-stage hierarchical models),第一階段使用一般加法模型(generalized additive models)波以松迴歸,分析個別區域單一及二空氣污染物對於肺炎及流感就醫風險。第二階段將個別區域估計單一污染物就醫風險係數,進行研究地區整合性就醫風險評估。此外,運用distributed lag model (DLM),評估單一及二空氣污染物/病毒陽性率/溫度/季節延遲效應對P&I就醫風險影響。空氣污染物分為單一日(single day)、三日移動(3 day moving average, 3d MA)平均濃度,污染物延遲效應從當日至前三日濃度(Lag 0-Lag 3),溫度延遲範圍從當日至前十三日。上述分析最終以每上升一單位污染物之濃度可能增加P&I就醫相對風險(relative risk, RR)來表示。 除了臭氧,大部分空氣污染物的當日濃度對P&I之就醫風險最大,且使用三日移動平均(3d MA)較單一污染物日平均估計之就醫風險高。 在單一及二空氣污染物分析,發現CO、NO2、NO、NOX、PM10、SO2,會使P&I就醫風險增加,而O3, 8h、O3, 24h的影響較不明顯。第二階段貝氏區域整合性評估中,發現污染物CO、NO2、NO、NOX、PM10及SO2與肺炎及流感就醫風險具有統計上顯著關係,相對危險分別為1.12(95%:1.01-1.23)、1.05(95%:1.02-1.08)、1.04(95%:1.001-1.07)、1.03(95%:1.00-1.049)、1.011(95%:1.002-1.02)、1.14(95%:1.03-1.25),而O3, 8h、O3, 24h於此分析中,未發現顯著相關性。在二污染物DLM分析,NO2累計4天(Lag 0-Lag 3)對於台北市P&I就醫RR值為1.131(95%:1.127-1.134),台北縣RR值為1.149(95%:1.146-1.152),桃園縣RR值為1.189(95%:1.184-1.193); 此外,在北北桃地區(不分區域)B型流感病毒陽性率與P&I門診就醫亦具有顯著相關性,每增加1%病毒陽性率則相對危險為1.0015(95%: 1.0009-1.0021)。

關鍵字

流感 肺炎 空氣污染 就醫 病毒

並列摘要


This study aim to evaluate the associations among air pollution, respiratory viruses infections and hospital admissions for pneumonia and influenza (P&I) in Taipei and Taoyuan, Taiwan. The insurance reimbursement claims in 1996-2007 from a randomly selected sample containing one million insured individuals were used to calculate the annual trend of daily clinic visits for cases on pneumonia and influenza (ICD-9 CM 480-486, 487). Virological surveillance data and atmospheric measurements (CO, NO2, NO, NOX, PM10, SO2, O3, 8h, O3, 24h) obtained from Centers for Disease Control and Environmental Protection Administration in Taiwan were also used. Poisson regression and Bayesian two-stage hierarchical models were applied to estimate the relative risks (RR) and 95% confidence intervals (CI) for hospital admission of pneumonia and influenza associated with air pollutants for each area and whole studied region. We found hospital admissions of P&I were significantly associated with CO, NO2, NO, NOX, SO2 with RR of 1.12 (95%:1.01-1.23), 1.05 (95%:1.02-1.08), 1.04 (95%:1.001-1.07), 1.03 (95%:1.00-1.049) and 1.14 (95%:1.03-1.25), respectively, for whole studied region by Bayesian two-stage hierarchical models. Daily influenza B virus also increased the risks of P&I with RR= 1.0015 (95% CI:1.0009-1.0021) per 1 % increase of positive isolation rate.

並列關鍵字

air pollution hospital admission influenza pneumonia virus

參考文獻


1. 行政院環境保護署. 地方環境資料庫. 行政院環境保護署; 2009.
27. Hwang JS, Chan CC, Hu TH. Air Pollution Mix and Emergency Room Visits for Respiratory and Cardiac Diseases in Taipei. Journal of Data Science. 2004;2:311-27.
62. Hwang Jing-Shiang , Hu Tsuey-Hwa, Chang-Chuan C. Air Pollution Mix and Emergency Room Visits for Respiratory and Cardiac Diseases in Taipei. Journal of Data Science. 2004;2:311-27.
2. Nicas M, Jones RM. Relative contributions of four exposure pathways to influenza infection risk. Risk Anal. 2009;29(9):1292-303.
3. Leigh MW, Carson JL, Denny FW, Jr. Pathogenesis of respiratory infections due to influenza virus: implications for developing countries. Reviews of infectious diseases. 1991;13 Suppl 6:S501-8.

延伸閱讀