臺灣博碩士論文加值系統

本研究利用2015年到2017年臺灣禽場資料與ebird鳥類資料，探討臺灣禽流感爆發和鳥的關聯性，透過列聯表分析探討數個危險因子；包括鳥的數量、鳥種數、Shannon多樣性指數、局部爆發之禽場比例等，再依之建立廣義線性模型以及廣義估計式。
針對不同的危險因子，透過尋求最大關聯性之切點，我們找出不同縣市與禽場爆發較相關之危險因子的勝算比，並且以所找出之切點當作將變數二元化的閥值。經由分析模型得知，臺南和嘉義在第三週與第四週時，鳥類和禽場的爆發為正相關，而雲林在第五週時，鳥類和禽場的爆發也為正相關。就較可信之第五至八週雲林和全臺灣的資料而言，由多變數廣義估計式之分析可發現禽場固定範圍內的爆發數與一些特定鳥種的出現與否是具有正相關的；反之，Shannon多樣性指數則呈現負相關。
我們將據此分析模型及結果初步定出一些危險因子的閥值，列出各縣市在不同時期與禽場爆發相關性較高的鳥種，期待對於未來禽流感疫情能提供以數據為基礎的監控機制。

This study investigated the association between the outbreak events of Taiwan avian influenza and bird-related risk factors, which were mainly ascertained through ebird data. The study period was from 2015 to 2017, and the association was explored primarily through contingency table analysis. Our risk factors included the total number of birds observed and reported through ebird system, number of species, Shannon diversity index, and the outbreak proportion within a limited realm circled around an index outbreak poultry farm. Generalized linear model (GLM) and generalized estimating equation (GEE) approach were utilized to explore the association systematically.
For various risk factors, cutoff points were selected towards maximal association. With these cutoff values, which were treated as thresholds for dichotomizing variables, we calculated the odds ratios between the outbreak events of Taiwan avian influenza and bird-related risk factors. Our models suggested that: for Tainan and Chia-Yi counties, the results exhibited positive association between outbreaks and bird counts in the first month (in particular the 3rd and 4th weeks). For Yun-Lin County, the positive association appeared at 5th Julian week. Restricted within 5th to 8th weeks, multivariate GLM regression using GEE method accounting for serial correlations revealed that several specific bird species were highly correlated with the outbreaks, while Shannon index gave negative results.
We suggested several threshold values for monitoring the risk factors of outbreaks appeared within specific time periods and specific counties. Substantial surveillance mechanism with regards to birds-related variables can be obtained by our models based on the data

摘要 i
Abstract ii
目錄 iv
表目錄 vi
圖目錄 vii
第一章　緒論 1
1.1 研究背景 1
1.2 研究動機 1
1.3 文獻回顧 2
1.4 研究目的與架構 3
第二章　研究方法 5
2.1　勝算比(odds ratio) 5
2.2 Shannon多樣性指數 6
2.3 兩個獨立母體比例的統計推論 6
2.4 廣義線性模型（generalized linear model，GLM） 7
2.5 廣義估計式（generalized estimating equation，GEE） 7
2.6 本論文所使用的統計模型及架構 8
第三章　禽流感爆發與鳥相關的危險因子 10
3.1 與禽場有關的基本資料: 描述與分析 10
3.1.1 禽場資料描述 10
3.1.2 爆發禽場基本分析 12
3.2 爆發禽場與鳥的數量之關聯性 15
3.2.1 禽場和鳥類數量 15
3.2.2 勝算比（Odd Ratio）之估計 17
3.2.3 尋找最佳切點 18
3.3 其他變數的勝算比及其切點 24
3.3.1 鳥種數 24
3.3.2 Shannon多樣性指數 26
3.3.3 區域性爆發比例 28
3.3.4 爆發比例 30
3.4 考慮剔除有爆發禽場周遭K公里之資料禽場 33
第四章 禽流感爆發與個別鳥種的關聯性分析 及統計模型 35
4.1 十五種高風險指標鳥種的勝算比 35
4.2 全臺灣和雲林的第五週都顯著鳥種 38
4.3 鳥種間有無出現列聯表 41
4.4 單變數廣義線性模型和廣義估計式分析結果 44
4.5 多變數廣義估計式分析結果 58
4.6 前一個月都沒爆發禽場 61
第五章　結論與建議 63
5.1 結論 63
5.2 建議 63
參考文獻 65
附錄 67

一、中文參考文獻
1.行政院農委會農業試驗所網站https://www.tari.gov.tw/
2.ebird網站https://ebird.org/home
二、西文參考書籍
1.McCullagh, P. and Nelder, J.A. (1989). Generalized Linear Models. 2nd edition, Chapman and Hall, London
三、西文參考期刊文獻
1.Alexander, D. J. (2000). "A review of avian influenza in different bird species." Veterinary microbiology 74(1-2): 3-13.
2.Cheng, M., et al. (2011). "The virulence variation in a H5N2 low pathogenic avian influenza virus after passage in 14-day-old chicken embryonic eggs." Taiwan Veterinary Journal 37(3): 221-226.
3.Fang, L. Q., et al. (2008). "Environmental factors contributing to the spread of H5N1 avian influenza in mainland China." PLoS One 3(5): e2268.
4.Gilbert, M., et al. (2006). "Anatidae migration in the western Palearctic and spread of highly pathogenic avian influenza H5N1 virus." Emerging infectious diseases 12(11): 1650.
5.Hulse-Post, D., et al. (2005). "Role of domestic ducks in the propagation and biological evolution of highly pathogenic H5N1 influenza viruses in Asia." Proceedings of the National Academy of Sciences 102(30): 10682-10687.
6.Kung, N. Y., et al. (2007). "Risk for infection with highly pathogenic influenza A virus (H5N1) in chickens, Hong Kong, 2002." Emerging infectious diseases 13(3): 412.
7.Lee, C. C., et al. (2014). "Emergence and evolution of avian H5N2 influenza viruses in chickens in Taiwan." J Virol 88(10): 5677-5686.
8.Lee, M.-S., et al. (2016). "Highly pathogenic avian influenza viruses H5N2, H5N3, and H5N8 in Taiwan in 2015." Veterinary microbiology 187: 50-57.
9.Liang, K.-Y. and S. L. Zeger (1986). "Longitudinal data analysis using generalized linear models." Biometrika 73(1): 13-22.
10.Lupiani, B. and S. M. Reddy (2009). "The history of avian influenza." Comp Immunol Microbiol Infect Dis 32(4): 311-323.
11.Margalef, D. R. (1957) "Information theory in ecology. (In Spanish.) " Mem. R. Acad. Barcelona 23: 373-449.
12.Pfeiffer, D. U., et al. (2007). "An analysis of the spatial and temporal patterns of highly pathogenic avian influenza occurrence in Vietnam using national surveillance data." The Veterinary Journal 174(2): 302-309.
13.Shannon, C. E. (1948). "A mathematical theory of communication." Bell system technical journal 27(3): 379-423.
14.Sims, L. D. (2007). "Lessons learned from Asian H5N1 outbreak control." Avian diseases 51(s1): 174-181.
15.Tam, J. S. (2002). "Influenza A (H5N1) in Hong Kong: an overview." Vaccine 20: S77-S81.
16.Thomas, M. E., et al. (2005). "Risk factors for the introduction of high pathogenicity Avian Influenza virus into poultry farms during the epidemic in the Netherlands in 2003." Prev Vet Med 69(1-2): 1-11.
17.Tiensin, T., et al. (2005). "Highly pathogenic avian influenza H5N1, Thailand, 2004." Emerging infectious diseases 11(11): 1664.