請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/84488
標題: | 台灣健保體制下,藥價、藥效與藥品費用分析 An Analysis of Drug Price, Effectiveness and Drug Expenditure under Taiwan National Health Insurance Scheme |
作者: | Yu-Shiuan Lin 林于瑄 |
指導教授: | 鄭守夏(Shou-hsia Cheng) |
關鍵字: | 健保,藥費,藥價,政策,驅動因子,比較性效果分析, National Health Insurance,drug expenditure,drug price,policy,comparative effectiveness analysis, |
出版年 : | 2022 |
學位: | 博士 |
摘要: | 研究背景 藥品費用佔台灣健保支出約 25-30%,遠高於其他有公營健康保險給付體系的國家。有鑒於此,健保持續以降低藥價及改變藥品給付架構等為主要策略,希望能減緩藥費上升的幅度,但台灣缺乏 2001 年以後藥費變化及影響因子的分析研究,難以評估藥品政策的效益及其影響。本研究第一部分欲統整 2001-2016 年健保藥品費用概況,以系統性和結構性的方法拆解健保藥費變化趨勢、探討藥費成長驅動因子及其影響程度。 降低藥價和推動學名藥是控制藥費的主要策略之一,但不同價格或廠別藥品的療效、安全性和處方型態可能不一致。本研究第二部分欲以真實世界數據為素材,從個人層級探討採用原廠藥和學名藥對治療效果及藥品相關費用的影響。 研究方法 本研究取 2001年、2006年、2011年、2016年健保申報數據,以三種現行可用的研究方法 (Laspeyres 方法 [Laspeyrex index, Y1998]、Fisher ideal 方法 [Fisher’s ideal decomposition, Y2004]、PMPRB方法 [相對影響力, Y2013]) 探討成年被保險人藥費成長驅動因子。描述性統計包括健保給付藥價變化趨勢、平均每人藥費、藥費佔率、學名藥使用率、新藥佔率等,並依世界衛生組織解剖治療及化學分類系統 (WHO/ATC system) 分類呈現。為能妥善驗證研究方法的可用性,比較不同研究方法的結果差異,本研究另選取四種特定疾病用藥進行深入探討:非胰島素降血糖用藥 (ATC code, A10B)、單一成份降血脂用藥 (ATC code, C10A)、全身性抗細菌用藥 (ATC code, J01)、抗腫瘤製劑 (ATC code, J01) 探究不同分析方法對研究結果造成的影響。計算藥品用量時,本研究導入標準日劑量 (standard daily dose, SDD) 進行同成分不同含量的商品換算,以進行合理累計。SDD 以 WHO/ATC system 定義日劑量 (defined daily dose, DDD) 為準,若該藥品缺乏可用 DDD,則參考原廠藥品仿單及實證醫學資料庫訂定之。 個人層級研究方面,本研究以 2011 年因骨髓炎和皮膚軟組織感染接受 glycopeptides 治療的住院病人為例,評估接受原廠藥或學名藥治療者,藥品累計劑量、住院日數、使用後線抗生素 (i.e., 注射劑型daptomycin、linezolid、tigecycline)、藥品及醫療相關費用是否存有顯著差異。 研究結果 2001-2016 年間,成年被保險人的健保藥費總計成長 2.59 倍。平均每人每年藥費和醫療費用分別由 5,231 元上升至 8,962 元、由 18,783 元上升至 31,135 元。新藥約佔研究年度藥費的 21.4-25.7%。學名藥佔率約為 40.6-48.1% (藥費) 和 83.6-84.7% (處方數)。 驅動因子分析結果顯示,整體藥費的主要驅動因子是藥品轉換 (drug switch),其次則為用量擴張。藥價降低減緩了藥費成長的速度,且 Laspeyres price index 低於健保公告藥價調降百分比中位數,顯示市場上選用藥品偏好調降價格較多者。特定疾病用藥分析指出,藥費驅動因子可能因藥品治療類別而不同。非胰島素降血糖用藥、單一成份降血脂用藥、抗腫瘤製劑的藥費驅動因子都以用量擴增為主、藥品轉換次之。抗腫瘤製劑的新藥對藥費影響較大,尤其是新診斷癌症病人。相對的,全身性抗細菌用藥可能因政策管制而限縮用量成長的幅度,而以藥品轉換為主要促進藥費上升的動力。 個人層級研究方面,研究結果發現接受學名藥治療者 (相對於接受原廠藥治療者) 似乎有累計劑量較高的傾向。因骨髓炎使用 teicoplanin 治療的病人,學名藥組藥品費用比原廠藥組略高,可能與累計藥品劑量較高有關。可惜符合條件的研究對象人數較少,統計檢定力略顯不足,以上結果未達統計學顯著差異。 研究結論 本研究建立台灣健保 2001-2016 年間藥費成長的趨勢,使用三種研究方法探討可能的藥費驅動因子,及不同治療類別藥品之間的差異。 研究結果顯示台灣總藥費及平均每人藥費比其他國家低廉許多。主要藥費驅動因子為藥品轉換,其次為用量擴張;但需注意不同治療類別藥品可能存有顯著差異。個人層級研究則指出,使用學名藥及原廠藥治療之累計劑量可能不同,降低了導入學名藥以控制藥費的政策成效;可惜研究檢力略顯不足。 Introduction Pharmaceutical expenditure represents a significant component of health care costs in Taiwan National Health Insurance (NHI). In recent decades, annual percentages of pharmaceutical spending to medical expense in Taiwan NHI were much higher than other countries with universal health care coverage. The first part of this study aimed to illustrate the growth trends of pharmaceutical expenditure during Y2001-Y2016 and identify its potential drivers. Drug price reduction and adopting generic medications are effective strategy of constraining drug expenditure. However, the equivalence of effectiveness, safety and prescription pattern between generic and brand-name agents are controversial. The second part of this study attempted to explore the utilization and effectiveness of generic and brand-name antibacterial agents using real-world data. Method The research conducted a retrospective population-based study focusing on adult insured using Taiwan NHI claim database at Calendar Year 2001, 2006, 2011, 2016. Descriptive analysis includes trends of list drug price change, drug expenditure per capita, percentage of drug cost to health care expense, proportion of generic drug utilization, percentage of new drug contribution to volume and expenditure, as well as drug utilization information of Anatomical Therapeutic Classification (ATC) groups. In order to explore the drivers of drug cost growth, prescription drug expenditure was decomposed by three methods: Laspeyres index (Y1998), Fisher ideal decomposition (Y2004) and PMPRB, Canada (Y2013). Additional to overall evaluation, four specific drug categories were examined in order to compare the efficacy of the decomposition methods, i.e., blood glucose lowering drugs exclude insulins (A10B), lipid modifiying agents, plain (C10A), antibacterials for systemic use (J01), antineoplastic agents (L01). Standard daily dose (complied based on WHO ATC/DDD; for drugs without available WHO ATC/DDD, the originators’ production information and evidence-based database were referenced for calculation) was applied for accumulated volume to minimize the effects of products with different delivery units. For the real-world comparative effectiveness analysis, patients administered to hospitals and received glycopeptides due to osteomyelitis or skin and soft tissue infection in Y2011 were selected. Measured indicators included accumulated dosage, length of stay, the use of antibiotics targeting vancomycin-resistant organisms, drug and health care expenses during index hospital stay and the following 1-year. Results During study period, total drug expenditure of adult insured increased 2.59-fold. Drug expenditure and health care cost per capita (adult) increased from NTD 5,231 to 8,962 and NTD 18,783 to 31,135, respectively. New drugs accounted for 21.4-25.7% of drug cost in study years. Generic drug adoption rates were 40.6-48.1% in expenditure and 83.6-84.7% in prescription. Drug switch was constantly the main drivers of total drug expenditure growth, following by volume expansion. Price reduction was beneficial for cost constriction. Laspeyres price index was usually lower than the median of list price change, which reveal that prescribers seems to favor drugs with reduced prices. Drivers of drug cost differed according to therapeutic category. Volume expansion was the dominant driver of drug expenditure growth of antidiabetics (exlude insulin), antihyperlipidemic and antineoplastic agents. The application of new drugs in newly diagnosed cancer patients also play an essential role in antineoplatic drugs. In contrast, drug switch (which may be respondent to policy control) was the leading cause of drug expense growth of systemic antibacterial agents. The real-world comparative effectiveness study demonstrated that patients treated with generic vancomycin or teicoplanin tended to receive higher dosage than those treated by brand-name comparators. Among patients with osteomyelitis treated by teicoplanin, drug cost seem to be higher in those receiving generic products than those reciving brand-name agents, which may be related to higher accumulated dose. Unfortunately, the difference did not reach statistical significance mainly because of inadequate study power. Conclusion The study established the trend of drug cost of Taiwan NHI during Y2001-Y2016. Three decomposition methods were applied to explore the drivers of drug expenditure growth and to demonstrate the difference between therapeutic categories. The results showed that drug price and drug expenditure per capita of Taiwan NHI remained much lower compared with other countries. Main drivers were drug switch and volume expansion. Real world evidence of comparative effectiveness revealed that patients treated by generic antibiotics tends to receive higher dosage than those treated by brand-name agents, but the study power was inadequate because of small sample size. |
URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/84488 |
DOI: | 10.6342/NTU202203880 |
全文授權: | 同意授權(限校園內公開) |
電子全文公開日期: | 2025-09-23 |
顯示於系所單位: | 健康政策與管理研究所 |
文件中的檔案:
檔案 | 大小 | 格式 | |
---|---|---|---|
U0001-2309202209343300.pdf 目前未授權公開取用 | 11.44 MB | Adobe PDF | 檢視/開啟 |
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。