隨著全民健保資料庫(NHIRD)相關研究的成長, 開發一個能進行世代分析研究的自動化平台是有必要的,本論文設計與實作一平台來分析疾病與藥物之間的關聯性,探討當罹患某種疾病進而服用某種藥物時, 會不會提高另外一種疾病發生的風險。我們蒐集了對此關聯性進行相關討論的研究論文,以驗證我們自動化平台的可行性。經過驗證後, 可發現我們所實作的平台確實可達到與驗證的論文結果有一致的方向性。;Objective: Currently none of analysis systems have considered the correlations between a drug and diseases. With the growth in number of data in the NHIRD (National Health Insurance Research Database), similar procedures, which have been used in previous cohort studies are performed again and again when carrying out similar cohort research and do the same procedures repeatedly is inefficient. In such circumstances an automatic analysis system ought to be able to reduce the laborious repeated procedures used in such circumstances. The aim of this study is to design and implement an automatic analysis system to analyze the correlations between a drug and diseases. Method: Based on NHIRD of the registration files and original claim data, we design and implement an automatic analysis system to analyze the correlations between a drug and diseases using cohort study design. We build a website to let users to enter their query. After each query is performed, our system transforms the query into corresponding SAS codes and then executes the SAS file by the statistical software until. Finally, our system converts the SAS output files into result tables and displays them on website. Results: We collect some cohort study researches from journals so as to verify the feasibility of our system, and input parameters by follow the previous researches parameters as equally as possible. Conclusion: Our system does make research more efficiently in using cohort study design to analyze correlations between a drug and diseases and save much more time for researchers.
