決策與資訊息息相關。因此有系統地整理蕪雜、繁複的政治資訊，乃屬刻不容緩的課 題。本文擬透過系統模型的分析，檢視政治資訊在政治系統傳輸運作的情形，並從而 析剖資訊系統的結構及其實際功能的運作過程。 首章導論，敘述研究動機與目的、研究方法與研究範圍；並略述啟發本文若干背景因 素（知識社會、知識工業的崛起）。 次章介紹資訊概念、資訊流程與資訊系統的結構。 第三章探討政治資訊系統。略述政治資訊與決策的交互作用，並從系統系絡析陳政治 資訊的傳輸過程、運作情形。最後述及政治資訊系統的結構。 第四章由系統生態特質的觀察與分析，敘述如何規劃及建立政治資訊系統。 第五章剖析發展中國家政治資訊系統的現況與改進之道。 第六章針對我國政治資訊系統的缺失，提出革新建議。 第七章結論與建議。全一冊，約七萬字，分七章，二十節。

藥物交互作用，輕則產生副作用或導致治療效果的降低，重者甚至會造成死亡或引發致命性的危險，這個問題確實值得引起大家的關注，尤其台灣醫師在開藥數目上較國外醫師來的多，引發藥物交互的機率也因而來的更高，如果能透過數據將此問題的嚴重性表露出來，且對此問題提出對策，對國內醫療品質的提升會是一大幫助。 本研究的資料來源為中央健保局資料庫，資料選定範圍為每年就診率最高的呼吸道疾病，想藉由統計分析和資料採礦方法分析，從這龐大的資料中，得到與藥物交互作用相關的有用數據，並將這些數據轉換成訊息，讓大家了解。 在統計分析的部分，從各個不同的角度來探討上呼吸道疾病的藥物交互作用現象，包括病患的基本資料，如：性別、年齡…，就醫行為，如：就醫月份、就分，開藥行為，如：開藥日份、開藥品項；至於 資料採礦分析的部分，則是利用C5.0決策樹，來找出高危險群；最後，針對整體中最容易產生高危險交互作用的藥物組，對其藥效、藥物機轉…等做說明。 / Drug interactions, to a lesser extent, could cause side effects or reduce the drug efficiency, to a greater extent, is life-threatening or could lead to fatality. This issue deserves the attention from the medical personnel and us, the users. The prescription practice in Taiwan more often than not gives out relatively high doses of medications than those of the practice in North America. Consequently the risk of having drug interactions increases respectively. If we could bring the attention of the related party into this alarming situation with the supporting statistical numbers, then hopefully we could set off the stage for a better medical treatment in Taiwan. We used the data from National Health Insurance database, selecting data only on the patients contracting respiratory system disease which sit for the largest portion in the database. We examine the data using statistical and data mining techniques in an attempt to extract the data on drug interactions and turn the number into useful information for the benefit of all related parties. In light of the statistical analysis, we have in depth analysis of the drug interactions on medications used in upper respiratory tract from different perspective such as patient’s record which includes age, the frequency in seeing the physician, details on drug prescription. As for the analysis, data mining technique used is C5.0 decision tree in determining the group at risk for getting the side effects of drug interactions. Moreover, we also have summarized the list of drugs that have shown the tendency to induce drug interaction so that extra caution should be taken when administering these groups of medications.

隨著國人之生活習慣改變與社會進步，糖尿病逐年上升，並成為國人十大死因之一，此外，糖尿病易引發其他疾病而產生併發症，當患者同時患有多種疾病，進而接受多重藥物治療，在使用多種藥物情況下易導致藥物產生交互作用，而中央健保局之「全民健康保險資料庫」囊括全體國人珍貴的醫療資料，其中包含詳細的糖尿病病患相關資料，本研究藉由分析中央健保局之糖尿病資料，目的在於分析糖尿病用藥中產生藥物交互作用之情形，希望能發掘有用資訊。 糖尿病資料中具有藥物交互作用處方佔有29％，藉由不同變數來探討糖尿病之藥物交互作用情形，譬如從年齡、性別、用藥藥品數、就醫科別等，進而得知資訊，諸如病患年齡與處方用藥數會影響是否有無藥物交互作用之產生；此外，藥物交互作用組合中，最多為口服低血糖用藥，這是糖尿病用藥，但產生最嚴重之藥物交互作用為心血管疾病用藥，並利用資料採礦技術挖掘有意義之規則。 / Incidence of diabetes mellitus is on the rise year over year in today’s modern living life style. It has been ranked as the tenth leading cause of death in Taiwan. When a patient suffers from more than a single cause of illness, he is treated by multiple medications which in turn could trigger drug interactions side effect. National health insurance research database has kept information containing data of patients with diabetes mellitus from which this research has based its data on. The purpose of this research is to analyze the side effects of drug interactions from diabetes mellitus medications. Of all prescriptions given for diabetes mellitus, 29% were resulted in having drug side effects. This research attempts to isolate the factor causing side effects of drug interaction of diabetes mellitus medications by difference variables, such as patient’s age, sex (male or female) and numbers of medicines. The techniques we employed in reaching our attempts is data mining.

當越來越多的藥物被發展出來進而治療人類的疾病，人們使用藥物時只知道該藥物對疾病的療效，卻忽略了其中可能隱藏的危機，不只是西藥，連中藥也隱藏相當大的危機。雖然國內有藥品管制局進行藥物交互作用的控管，但是藥物交互作用真的被有效控管嗎？成效又如何呢？有鑑於此，這成為大家都想知道的問題。 利用健保資料庫中龐大的門診資料，嘗試推估國人藥物交互作用的情形，由國人十大死因中得知，其中尤以慢性病病患的情形更為嚴重，例如：惡性腫瘤、腦血管疾病、心臟疾病、糖尿病、肝病和高血壓，本次研究以心血管病患為討論主體，試瞭解在長期服用心血管藥物時，醫生在已知病患病情之下卻無法避免藥物交互作用的情形為何，利用資料採礦的方法，找出可能造成交互作用發生的原因、族群，進而去提醒醫療機構避免其發生。 研究發現，心血管疾病病患分為一些併發症族群，如：心臟病、心血管疾病糖尿病、關節炎、消化潰瘍，從這些病患用藥中發現許多交互作用的用藥與其併發症也許多相關。高血壓疾病病患大多以強心劑、利尿劑為主的交互作用，心血管疾病併發糖尿病病患大多以口服降血糖為主的交互作用，心血管疾病併發關節炎病患大多以解熱鎮痛劑為主的交互作用，而心血管疾病併發消化潰瘍病患大多以制酸劑為主的交互作用。 健保局應當面對且對交互作用之高危險族群作一共同用藥規則，將用藥危機降到最低，讓心血管病患在治療的過程中，得以享有良好的醫療品質，亦不造成藥物濫用、浪費之情形。 / As more medicines are being developed to cope with diseases, most of the users think of the therapeutic effect of the medicine without thinking of the risk that might have associated with drug interactions. The risk of drug interactions could have set chemical reactions thereby causing drug side effects. Although the National Bureau of Controlled Drugs has set a procedure for controlling drug interactions, the issue of validity and efficiency remains an open question. The trend of people contracting chronic diseases is on the rise, one of which is cardiovascular. The attempt of this paper is to observe for the effect of drug interactions if any for the long term usage of the cardiovascular medications under the supervision of the doctors. We adopt data mining techniques to single out the probable variable causes in triggering negative effect of drug interactions and presented them to related medical personnel so that a tightening measure is adopted when administered the medications. The data used in this research comes from the National Health Insurance Research Database. Our research findings have reviewed that cardiovascular disease patients suffering from the complications, such as diabetes, arthritis, peptic ulcer and hear disease are highly related to the drug interactions. Heart disease patients are at risk of cardiac stimultants and diuretics. Diabetic’s patients taking the Sulfonylureas suffer from its interactions. And arthritis patients are at risk of having the side effects of taking Aspirin. Bureau of National Health Insurance should set a standard procedure in monitoring the prescription given by the medical personnel in an effort to reduce the risk of drug interactions and set off the stage of quality medical treatment to keep off the abuse of drug usages.

科学研究費補助金 研究種目:基盤研究(B) 課題番号:17310151 研究代表者:金井 篤子 研究期間:2005-2007年度

京都大学 / 新制・課程博士 / 博士(医学) / 甲第23416号 / 医博第4761号 / 新制||医||1052(附属図書館) / 京都大学大学院医学研究科医学専攻 / (主査)教授 柳田 素子, 教授 木村 剛, 教授 近藤 尚己 / 学位規則第4条第1項該当 / Doctor of Medical Science / Kyoto University / DFAM

近年來，基因與疾病之關聯分析 (association analysis) 越來越受到研究學者重視，因為在複雜性疾病與易感性基因之探討中 傳統的連鎖方法 (linkage method) 已不適用，所以複雜性疾病與易感性基因的關聯分析也蓬勃發展起來。在本文中我們主要是在探討 關聯分析中以家庭為研究資料與以群體為研究資料之間的優缺點，進而取長補短提出結合兩種資料之新的關聯分析方法 來增加估計與檢定之效力。我們同時考慮環境因素，探討基因因素與環境因素之交互作用。 本研究共分為三部份。第一部份探討如何整合病例-父母/病例-同胞 (case-parent/case-sibling) 與病例-對照 (case-control) 研究。我們提出一個加權最小平方 (Weighted Least Squares) 的方法將病例-父母/病例-同胞與病例-對照分析之估計式加以結合，以增進統計檢定之效力。 第二部分旨在探討基因-環境之交互作用。我們提出一個二階段研究設計法。在第一階段研究中，先收集病例資料; 在第二階段研究中，再收集其相對應之控制組資料。我們提出一個迴歸估計式以結合第一階段之單純病例分析(case-only analysis) 與第二階段之病例-對照分析。此建議之估計式即使在基因因子與環境因子 獨立之條件 (此條件為單純病例分析所必需) 不成立的情形下，依然可得出正確之統計推論。 第三部份旨在探討群體分層 (population stratification) 存在 之情形下，基因-環境之交互作用。我們提出一個二階段研究設計，以病例資料為第一階段資料， 再從病例資料中隨機抽取一部份病例患者之父母資料為第二階段資料。我們提出一個迴歸估計式結合單純病例研分析與病例-父母分析之估計式。 此新估計式即可整合單純病例分析與病例-父母分析，同時在群體分層存在之情形下，仍可得出有效之統計推論。 / In recent years, there are increasing attention to association studies, because linkage method will not be suitable under complex disease and susceptible genes. In the thesis, we are probing into association of family study and population study. And we combine family study and population study for increased efficiency of association method. We also consider interesting studies about gene-environment interactions. The thesis contains three projects. The first project focuses on examining when and how the two sources of information offered by such studies, one from the case-parent/case-sibling analysis, and the other from the case-control analysis with data from affected subjects and unrelated controls, can be integrated to enhance statistical power. We propose a weighted least-squares approach to linearly and optimally combine separate estimators from the case-parent/case-sibling and the logistic regression analysis for the association parameters. In the second project, we focus on examining the situation of gene-environment interaction. We propose a two-stage design. In the first stage, we collect patient data, and we seek out control data with respect to cases in the second stage. We propose regression analysis estimation in order to combine the case-only analysis in the first stage and the case-control analysis in the second stage. This estimation earns the correct statistical inference when genes and environment factors are not independent. In the third project, we explore gene-environment interactions under population stratification. We propose a two-stage design. In the first stage, we collect patient data, and we randomly collect a partial data of patient's parent from the cases in the second stage. We propose regression analysis estimation in order to combine the case-only analysis and the case-parent analysis. This estimation can combine the case-only analysis and the case-parent analysis, and attains effective statistical inference under population stratification.