1999至2005年台灣地區腸病毒感染的季節變動

中山醫學大學機構典藏 CSMUIR > 健康管理學院 > 公共衛生學系暨碩士班 > 博碩士論文 > Item 310902500/872

Please use this identifier to cite or link to this item: https://ir.csmu.edu.tw:8080/ir/handle/310902500/872

Title:	1999至2005年台灣地區腸病毒感染的季節變動 Seasonal Variation of Enteroviruses Infection in Taiwan From 1999 to 2005
Authors:	林敏琮 Min-Tsung Lin
Contributors:	中山醫學大學：公共衛生學系碩士班;李鴻森
Keywords:	腸病毒;季節變動;繪圖;卡方適合度檢定;分解法;圓形分佈 enterovirus;seasonal variation graphics chi-square test for goodness of fit;decomposition;circular distribution
Date:	2007/06/28
Issue Date:	2010-03-22T04:30:20Z (UTC)
Abstract:	研究背景：傳染病的發生普遍具有季節變動的現象，瞭解傳染病的季節性變動將有助於預防工作的推行。國外已有長期監測的研究報告指出，腸病毒的感染發生具有季節性，不過這些研究對於分析季節變動的方法，普遍未有太多著墨。考量統計理論與分析工具的不同，有助於研究者對資訊的掌握。因此本研究的目的是使用官方監測腸病毒的資料描述腸病毒在臺灣的發生情況，並採用不同的統計方法，分析腸病毒發生的季節變動趨勢。材料與方法：使用之研究材料為1999年至2005年臺灣腸病毒的監測資料，包括定點醫師於每週報告之手足口病與疹性咽峽炎病例、法定傳染病中應報告的腸病毒感染併發重症確定病例與病毒性感染症合約實驗室之腸病毒感染檢驗陽性病例，分別應用繪圖、卡方適合度檢定、分解法與圓形分佈法進行分析。結果：應用繪圖直接呈現三種監測資料的季節性分別為春夏季節交替之際與秋天。季節高峰月份以卡方適合度檢定分析結果，分別為定點醫師監測、法定傳染病個案通報與病毒合約實驗室資料均為非分均勻分布（p<0.001）。分解法分析的結果分別是定點醫師監測的高峰為五-六月（季節指數分別為2.38與2.01）、法定傳染病個案通報為五-六月與十月（季節指數分別為1.75、1.62與1.45）、病毒合約實驗室為五-六月與十一月（季節指數分別為2.1、1.6與1.1）。圓形分佈所求得的季節高峰月份，分別是定點醫師監測為六月（168.8°，p<0.001）、腸病毒重症為七月（193.2°，p<0.001）、合約實驗室為六月（178.8°，p<0.001）。結論：本研究分別應用繪圖、卡方適合度檢度、分解法與圓形分佈法等四種分析方法，皆得到臺灣的腸病毒感染有季節變動趨勢的結果，而季節高峰分別為春夏季節交替之際與秋天。分析季節性時，若資料型態為長期資料且為頻繁出現事件，建議可使用分解法，除了季節變動趨勢以外，還可瞭解趨勢與循環的型態。若僅想呈現季節性趨勢，可使用繪圖配合卡方適和度檢定，避免流於直覺的主觀判斷。若由繪圖後發現該事件具有單一時間集中趨勢，則可使用圓分佈法呈現其集中趨勢的結果。 Background Most of the communicable diseases occurred with seasonal variation. Understanding this seasonal pattern would contribute to the disease prevention work. Several studies have suggested the seasonal variation of enterovirus infection. However, few studies have focused on comparing the methods of analyzing the seasonal variation. Different statistical methods based on different theories would help the researcher to manage the various types of communicable disease data. The purposes of this study were to describe the epidemiology of enterovirus infection in Taiwan and to discuss the utility of different statistical methods in analyzing seasonal variation of communicable diseases. Method The study needs enterovirus infections data in Taiwan from 1999 to 2005. The sources of data include the sentinel surveillance system, the Notifiable disease reporting system and the contract lab data. Statistical methods used for the analysis of seasonal variation include graphics, chi-square test for goodness of fit, method of decomposition and circular distribution. Results The findings of graphics show clear seasonality in late spring-early summer and autumn of all the three types of surveillance data. The chi-square test for goodness of fit indicates that the sentinel surveillance, Notifiable disease, and the contract lab data are not uniformly distributed (p<0.01). By decomposition method, the sentinel surveillance peaks were in May and June (seasonal index was 2.38 and 2.01), the Notifiable disease peaks were in May, June and October (seasonal index was 1.75, 1.62 and 1.45), and the contract lab peaks were in May, June and November (seasonal index was 2.1, 1.6 and 1.1). By circular distribution, the sentinel surveillance peak was in June (mean angle was 168.8° and p<0.001), the Notifiable disease peak was in July (mean angle was 193.2° and p<0.001), and the contract lab peak was in June (mean angle was 178.8° and p<0.001). Conclusion In this study, we presented the seasonal variation patterns of enterovirus infections in Taiwan by four statistical methods. It was suggested that the decomposition method be used to analyze long-term data to obtain the pattern of trend-cycle and seasonality. Graphics and chi-square test for goodness of fit could present the trend of seasonality and avoid subjective judgment. When graphics shows peak occurrence in single month, the circular distribution could be used further to explore the central tendency.
URI:	http://140.128.138.153:8080/handle/310902500/872
Appears in Collections:	[公共衛生學系暨碩士班] 博碩士論文

Files in This Item:

File	Description	Size	Format
林敏琮.pdf	碩士論文	908Kb	Adobe PDF	1283	View/Open

Loading...