本文研究慢性病治療記錄卡用藥管理作業，所需之相關資料庫的設計與建置。由記錄歸戶作業開始，建立一套可用之「慢性病治療記錄資料庫」。在慢性病治療記錄資料庫中，整理、分析慢性病病人，在各醫療院所，不同科別就診時，及最近服用的各種慢性病貴重和禁忌管控藥品記錄。定期地歸戶與分析慢性病治療記錄資料庫，可供醫師及個人對慢性病治療過程的參考。

　　本文所建置之簡易用藥成份交互作用資料庫，除了在每次就診時，即時地提供看診醫師正確用藥資訊參考之外，亦可以在事後的歸戶作業中，分析個人用藥狀況，以及用藥嚴重交互作用之情形。

　　本文採用健保局南區分局所提供，自1999年起三年內南區分局所轉存慢性病患，高血壓組192人與糖尿病組43人，施行歸戶建檔、用藥重複檢查，以及用藥交互作用檢索，其結果統計如下：糖尿病群組每個人平均重複用藥處方135次，每人平均浪費22,449元；高血壓群組每個人平均重複用藥處方120次，每人平均浪費21,813元。就整體而言，糖尿病群組重複用藥的筆數比例為26.29%，浪費金額的比例為20.62%；高血壓病人群組重複用藥的筆數比例為26.03%，浪費金額的比例為22.08%。而在用藥交互作用檢索發現，糖尿病群組43人，總用藥成份筆數2,666筆，找到用藥嚴重交互作用0筆；高血壓群組192人，總用藥成份筆數12,556筆，找到用藥嚴重交互作用筆數7筆。

　　This thesis is focusing on the studies the chronical patient’s medical administration and implementation of the drugs management database for the proposed chronic disease patient data card. Starting from the post-clustering algorithm, chronic disease patient’s records has been implemented a feasible archives. Then the patient's drug records, such as precious and taboo medicine prescribed at every medical clinic and various hospital departments, can be summarized and analyzed for personal chronically medical contradicted archives. Regularly clustering and analyzing the chronic disease drug records archives, both the doctors and the patients can take it for consultation.

　　According to the simple database of drugs serious interaction implemented by this study, the doctor can not only access the correct medicine information for chronical disease in real time. But also can obtain the medication statistics of the serious drugs interaction for each patient by post-clustering algorithm.

　　According to drugs consumption sample data of 192 high blood pressure (HBP) patients group and 43 diabetes patients group collected from the National Health Insurance (NHI) in the three years. By undertaking through the checking of repeated medicine drugs and serious drugs interaction, each diabetes patient repeating 135 drug prescription and wasting 22,449NT$；each HBP patient repeating 120 drug prescription and wasting 21,813NT$. On the whole, the diabetes group has 26.29% repeated medicine drug prescriptions, and has wasted 20.62% of the total medicine cost. The HBP group has 26.03% repeated medicine drug prescriptions, and has wasted 22.08% of the total medicine cost. As for the serious drugs interaction searching no serious interaction of 2,666 medical prescriptions for 43 diabetes patient group but 7 serious interaction of 12,556 medical prescriptions for 192 HBP patient group.

【1】”IC卡的控制卡晶片”，零組件雜誌62期，P113-P115，1996年12月。

【2】王國榮，”Visual Basic 6.0 資料庫程式設計”，旗標出版股份有限公司，2001年。

【3】林璨，”藥費成長原因分析及制度面改革方向”，中央健保局醫務管理處，1999。

【4】彭懋芝，”智慧卡應用系統之安全性探討”，電腦與通訊，第64期，1997年11月。

【5】溫家麟，”二代醫療網之醫師卡安全作業之研究”，國立成功大學工程科學研究所碩士論文，2001年6月。

【6】樊國楨，”電子商務高階安全防護:公開金鑰密碼資訊系統安全原理”，資策會圖書，1997年。

【7】賴溪松、韓亮、張真誠，”近代密碼學及其應用”，松崗，1998。

【8】賴俊澤，”二代醫療網之健保IC卡安全作業之研究”，國立成功大學工程科學研究所碩士論文，2001年6月。

【9】Berry, M. J. and Linoff, G. “Data Mining Techniques for Marketing, Sales and Customer Support.”, Wiley, 1997.

【10】“Bull Smart Kit Cards and Security Module Contents Description Guide”,CD ROM—Crypto BuidlerBullSB DocTBMap.pdf,Bull CP8 Inc. ,1998.

【11】“Bull TB Cards Service Provider User Guide”,CD ROM—Crypto BuidlerBullSBDocTBCSP.pdf, Bull CP8 Inc. ,1998.

【12】Deming, W. E. (1943). Statistical Adjustment of Data, Wiley.Friedman, J. “Data mining and statistics: What is the connection?”, The 29th Symposium on the Interface, Houston, TX, May 1997.

【13】Henry Dreifus, J. Thomas Monk, “Smart Cards—A Guide To Building and Managing Smart Card Applications”, Wiley Computer Publishing,1998.

【14】ISO 7816,“Identification cards-Integrated circuit(s) cards with contacts”,International Organization for Standardization.

【15】J.P.Griffer,“A Manual of Adverse Drug interactions”, John Wright, 1988.

【16】M.J.A. Berry & G. Linoff , “ Data Mining Techniques, JohnWiley & Sons”, 1997.

【17】”中央健保局”，http://www.nhi.gov.tw/02hospital/hospital_file/fchi05-a.exe

【18】”成醫圖書館Medline”，http://www.medlib.ncku.edu.tw

【19】”Java CardTM2.1.1 Technology Datasheet”，http://java.sun.com/products/javacard/datasheet.html

【20】”Oberthur”，http://www.oberthur.com/