請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/69673
標題: | 新上市藥品新使用者研究設計:以抗凝血劑作為範例 Modified Prevalent New User Design: An Illustration with Oral Anticoagulants |
作者: | Hui-Min Lin 林卉敏 |
指導教授: | 陳建煒(K.Arnold Chan) |
關鍵字: | Prevalent new user design,暴露集合,傾向分數配對, Prevalent new user design,exposure sets,propensity score matching, |
出版年 : | 2018 |
學位: | 碩士 |
摘要: | 研究背景:
新藥於上市後須積極監測安全與療效,因此需要設計完善的藥物流行病學研究協助了解一般臨床藥品使用與治療情況。過去流行病學研究使用的new user design和prevalent user design各有優缺於觀察新上市藥品之療效與安全性。臨床上,若病人為診斷疾病後使用新上市藥品,這些藥品使用者將稱之為新藥new user。另有病人於新藥上市前即被診斷疾病而使用了一陣子現有治療並在新藥上市後才改使用新藥,此類病人即視為新藥switcher。本研究改良並使用新研究方法prevalent new user design於抗凝血劑藥品為一範例,作為新上市藥品研究設計之討論。 研究方法: 本研究使用2011年至2015年台灣全民健康保險資料,並且以新發非瓣膜性心房震顫病人成立初步研究世代。再因應當病人開始使用新藥dabigatran處方時建立一個暴露集合(exposure set),並納入warfarin處方使用者擁有與dabigatran使用者類似之過往抗凝血劑經驗於暴露集合中。利用傾向分數之計算,於集合中配對出傾向分數最近的dabigatran和warfarin處方,並以該處方使用者建立出研究世代。最後於此研究世代探討腦出血、腸胃道出血、栓塞性中風和死亡風險比較。 研究結果: 初步研究世代包含30,168位心房震顫病人,並在暴露集合和傾向分數配對設計後納入了22,334位追蹤對象於研究世代。相較於warfarin使用者,dabigatran使用者有較低的腦出血(相對危險(HR)=0.50; 95% 信賴區間(CI): 0.39, 0.65)、腸胃道出血(HR=0.77; 95%CI: 0.67, 0.88)以及死亡風險(HR=0.58; 95%CI: 0.54,0.63)。而對於栓塞性中風兩者風險無異。 結論: Prevalent new user 研究設計目標為在新藥藥品上市時盡量納入所有新藥藥品使用者包含new users和switchers。利用暴露集合之設計更是能針對switchers控制於新藥使用前之其他藥品暴露所造成的影響並且偵測過去其他藥品使用時發生的研究事件。總結,此方法可提供新上市藥品更為精緻的安全與療效分析亦可提供不同型態新藥品使用者之分析結果。 Background: Timely safety and effectiveness outcome evaluation for new drugs were expected from well-designed pharmacoepidemiolgy studies. New user design and prevalent user design each shared its shortcomings especially when studying recently launched new drugs. In clinical practice, patients offered a new drug once being diagnosed would be considered as a new drug new user. Whereas patients treated with an old drug and switched to the new drug afterwards would be considered as a new drug switcher. Thus an innovative study design, prevalent new user, was suggested to study new drugs encompassing both new users and switchers. We implemented the design with modification and demonstrated study design concerns in practice on a novel drug, dabigatran, among oral anticoagulants for illustration. Methods: An incident non-valvular atrial fibrillation base cohort was constructed in this study utilizing Taiwan National Health Insurance data from 2011 through 2015. When a patient initiated dabigatran, warfarin prescriptions with comparable prior oral anticoagulant to the dabigatran initiating prescription would be included in an exposure set. Propensity score (PS) estimated within the exposure set for the dabigatran and warfarin prescriptions were matched to select the dabigatran and warfarin users to be included in the study cohort. Intracranial hemorrhage, gastrointestinal bleeding, ischemic stroke and all-cause mortality were evaluated among patients in the study cohort. Results: There were 30,168 patients included in the base cohort. After constructing exposure sets and PS matching, 22,334 patients were included in the study cohort for safety and effectiveness outcome evaluation. Decrease in risk of intracranial hemorrhage (hazard ratio(HR)=0.50; 95% confidence interval(CI): 0.39, 0.65), gastrointestinal bleeding (HR=0.77; 95%CI: 0.67, 0.88) and all-cause mortality (HR=0.58; 95%CI: 0.54,0.63) were observed among dabigatran prevalent new users comparing with the matched warfarin users. Similar effectiveness in stroke prevention were observed among dabigatran and warfarin users. Conclusion: Prevalent new user design intended to incorporate sizeable new drug users, including both new users and switchers, soon after a drug was introduced. Exposure sets applied were to control for previous drug use experience as well as taking into account of events occurred during the prior drug exposure period among switchers. As a result, refined analyses were carried out among dabigatran users as well as estimates among new users and switchers. |
URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/69673 |
DOI: | 10.6342/NTU201800934 |
全文授權: | 有償授權 |
顯示於系所單位: | 流行病學與預防醫學研究所 |
文件中的檔案:
檔案 | 大小 | 格式 | |
---|---|---|---|
ntu-107-1.pdf 目前未授權公開取用 | 1.53 MB | Adobe PDF |
系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。