傳染性疾病傳播之空間分析與時空分析之適切性比較：以2002年底特律之西尼羅病毒疫情為例

A comparison for appropriateness of spatial analyses and spatio-temporal analyses: An example of the West-Nile virus epidemic in Detroit, 2002.

林惠君(Lin Hui Chun)

指導教授 ： 李正宇

亞洲大學/資訊學院/生物資訊學系碩士班/碩士(2008年)

摘要

全球氣候的轉變和交通的便利使得傳染性疾病(Infectious diseases)的傳播變得容易且快速，且其主要發生範圍亦產生遷移；因應此種變化，各先進國家都在找尋、嘗試各種方法，希望能對疾病傳播有更準確的分析及預測。本研究的目的是希望藉由比較純空間分析（Spatial Analysis）及時空分析(Spatial-Temporal Analysis)的結果來找出分析傳染性疾病傳播的較佳方法。此研究以2002年夏天在美國密西根州爆發在烏鴉及人類間的西尼羅病毒（West-Nile virus）分別進行空間分析（採用空間分析軟體Passage v1.1）及時空分析（採用時空分析軟體IEAST v1.3）作分析結果比較。由此結果發現純空間分析方法隱藏或混淆了原疾病傳播統計現象中的時空關聯性(space-time correlation)，導致產生一些誤導研究人員的結果，進而影響對疾病蔓延的控制及預防策略。此研究的結果建議，為了真實表達疾病傳播之統計行為並避免產生誤導的分析結果，對於傳染性疾病分析的傳播模型應採取時空模型作為分析的基礎，而非使用單純時間分析或單純空間分析。

關鍵字

傳染性疾病 ； 空間分析 ； 時空分析 ； 時空關聯性

並列摘要

Global climate change and better transportation make infectious diseases spread much more easily and faster as before. Besides, traditional coverage of certain diseases is still changing. Coping with these changes, many countries are looking and searching for potential methods, and hope to conduct analyses with higher accuracy and better forecasts for the spreading. The main objective of this research is to compare the analysis results from pure spatial analysis and spatial-temporal analysis for a given space-time dataset, and to find the differences between the inferences from these analyses. We attempted to give a supporting evidence for the better method in the analyses of disease epidemics.
The space-time dataset used for the research was from a West Nile Virus outbreak among crows and human in Greater Detroit (Michigan, US) during the summer 2002. The results of the analyses conducted using both spatial analysis (using a spatial analysis software -- PASSAGE v1.1) and STARMA-based (space-time autoregressive moving-average models) spatial-temporal analysis (using spatio-temporal analysis software IEAST v1.3) were interpreted and compared. In the comparison, we found that using pure spatial analysis method may conceal or confuse the facts of the space-time behaviors in the statistical processes of the disease transmission. It is possible to cause misleading inferences, and thereby to deteriorate the control and prevention of diseases. This research strongly suggested that, to avoid misleading analyses and to correctly portrait the statistical behavior of diseases spreading, spatio-temporal models and methods should be used for analyses of infectious disease spreading, rather than pure spatial or temporal models or methods.

並列關鍵字

Infectious diseases ； Spatial Analysis ； Spatial-Temporal Analysis ； space-time correlation ； Space-Time Autoregressive Moving-Average Models ； STARMA

參考文獻

19. 胡立諄、賴進貴，臺灣女性癌症的空間分析，台灣地理資訊學刊 (4):39-55,2006.

2. David J. Rogers, Sarah E. Randolph, Robert W. Snow and Simon I. Hay.Satellite imagery in the study and forecast of malaria. nature 415 (7),2002.

3. Emma K. Chaput, James I. Meek, and Robert Heimer.Spatial Analysis of Human Granulocytic Ehrlichiosis near Lyme, Connecticut. Emerging Infectious Diseases.8(9), 2002.

Using autoregressive integrated moving average (ARIMA) models to predict and monitor the number of beds occupied during a SARS outbreak in a tertiary hospital in Singapore. BMC Health Services Research,2005.

5. Farzad Mostashari, Martin Kulldorff, Jessica J. Hartman, James R. Miller, and Varuni Kulasekera.Dead Bird Clusters as an Early Warning System for West Nile Virus Activity.Emerging Infectious Diseases.9(6),2003.