Please use this identifier to cite or link to this item: https://ah.lib.nccu.edu.tw/handle/140.119/56425
題名: 全民健保部分負擔制度之研究
作者: 連賢明;許績天;陳素秋
貢獻者: 國立政治大學公共行政及企業管理教育中心
行政院衛生署
關鍵詞: 部分負擔;道德風險;醫療利用
日期: 2009
上傳時間: 7-Dec-2012
摘要: 自全民健保開辦以來,為維持健保財務穩健,健保局過去十多年間八次調整部分負擔看出。以醫學中心為例,一般門診部分負擔金額由100元逐次上升至360元,調整幅度之大,可見一般。透過提高部分負擔,健保局一方面透過增加健保收入,一方面則希望透過調整部分負擔來抑制被保險人不當醫療使用(亦即健康保險中的「道德風險」問題)。但提高部分負擔也引起外界許多批評,擔心降低民眾(特別是弱勢族群)就醫的需求,對就醫公平性及可近性皆造成更大傷害。究竟調整部分負擔僅是抑制醫療利用或是造成就醫的不公平?這個答案取決於就醫時的價格彈性。由於全民健保下所有政策調整一體適用,僅能就部分負擔政策變動前後相對應的醫療利用變化來推估價格效果。這類分析由於僅利用「實驗組」,缺乏「對照組」的比較,造成可能估計偏誤。因此,本計畫從透過三方面來進行價格彈性估計,並以所得價格效果,來評估政策調整方向。首先,我們使用健保歸人檔分析。健保歸人檔涵蓋120萬人超過十年的醫療利用資料,可使我們能長期針對不同醫療服務、不同年齡民眾來分析觀察;其次,我們從免部分負擔樣本中,利用配對估計法(propensity score matching,PSM)挑選控制組,使控制組與實驗組樣本相近,再以差異中差異方法進行分析;最後,為了納入部分負擔中不同醫院層級間金額有所差異,我們預計使用機率模型(logit model)來估計到醫院就診的機率,或是巢狀邏輯特性模型(nested logit model)來涵蓋不同層級醫院選擇,以納入不同層級間醫院部分負擔價格不同的問題。
In order to maintain the financial stability, the Bureau of National Health Insurance (BNHI) raised the co-payment scheme eight times in these 10 years since the inauguration of Taiwan’s National Health Insurance in 1995. For example, the co-payment for an outpatient visit in a medical center is gradually raised to NT$360 from NT$100. By raising the co-payment, the BNHI hopes to increase the premium on the one hand and eliminates the inefficiency arising from moral hazard problems on the other hand. Nonetheless, the increase casts a lot of doubts because the co-payment might be detrimental to the health equality and accessibility of medical care. To what extent the health inequality is affected, however, depends on the corresponding price elasticity of health care. Because the policy adjustment applies to all individuals in Taiwan, we can only evaluate those price effects by comparing prior and after the change of co-payments. However, this kind of comparison is likely to introduce endogenous bias since we have only “treatment groups”. To attenuate the bias, this project evaluates the price effects through three dimensions. First, we investigate the change of health care utilization patterns using the panel data consisting of 5% or 1 million NHI enrollees in the past ten years. This data can be used to investigate the long term pattern for varies medical treatments and patients with different ages. Secondly, in order to solve the selection biases, we match match patients of treatment groups with that in one million random sample using propensity score matching. We then use patients in “quasi-control group”(from PSM) and “treatment group” and estimate the price elasticity by difference-in-difference (DID) method. Lastly, in order to separate the effects due to the different co-payment charged by hospitals of different accreditations, the Logit model and nested Logit model will be used to control for the potential influences on estimations.
關聯: 基礎研究
委託研究
研究期間:9806~ 9812
研究經費:780仟元
資料類型: report
Appears in Collections:研究報告

Files in This Item:
File SizeFormat
RRPG98070058.pdf1.26 MBAdobe PDF2View/Open
Show full item record

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.