請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/7286
標題: | 模擬臺灣推行手術服務量閾值之影響-以五種手術為例 Simulating the impact of implementation of volume threshold in Taiwan- using five surgeries as examples |
作者: | Ying-Yi Chou 周盈邑 |
指導教授: | 董鈺琪 |
關鍵字: | 服務量,閾值,照護結果,工具變項,成本效果, volume,threshold,outcomes,instrumental variables,cost-effectiveness, |
出版年 : | 2019 |
學位: | 博士 |
摘要: | 研究背景與目的:目前已有研究針對全膝關節置換術、全髖關節置換術、冠狀動脈繞道手術、頸動脈支架置入以及二尖瓣置換與修復術,探討服務量與結果關係,然而少有研究證實上述手術是否存在服務量閾值。此外,目前較少研究控制未觀察到之因子對服務量與結果關係之影響。最後,目前尚未有研究探討實施醫院服務量閾值之成本效果。因此,本研究有三個目的:(一)針對全膝關節置換術、全髖關節置換術、冠狀動脈繞道手術、頸動脈支架置入,以及二尖瓣置換與修復術,探討醫院與醫師服務量閾值;(二)針對全膝關節置換術、全髖關節置換術、冠狀動脈繞道手術、頸動脈支架置入,以及二尖瓣置換與修復術,探討醫院及醫師服務量閾值與結果之關係;(三)針對全膝關節置換術、全髖關節置換術、冠狀動脈繞道手術、頸動脈支架置入,以及二尖瓣置換與修復術,探討推行醫院服務量閾值之成本效果。
研究方法:針對研究目的一與二,資料取自衛生福利資料科學中心全民健保資料檔、死因統計檔等,研究對象為2015年接受全膝關節置換術與全髖關節置換術之病人、2014年至2015年接受冠狀動脈繞道手術之病人,以及2011年至2015年接受頸動脈支架置入與二尖瓣置換與修復術之病人。本研究以限制性立方截斷式(restricted cubic splines)模型、接受器操作特性曲線(receiver operating characteristic, ROC)以及Youden指數,探討醫院與醫師服務量閾值,並以廣義估計方程式與工具變項,探討醫院與醫師服務量閾值對照護結果與醫療利用之關係。針對研究目的三,資料取自上述資料庫與過去文獻,研究對象為2012年接受全膝關節置換術與全髖關節置換術之病人,以及2013年接受冠狀動脈繞道手術、頸動脈支架置入,與二尖瓣置換與修復術之病人。本研究以馬可夫模型評估實施醫院服務量閾值之成本以及健康生活品質校正生命年(quality-adjusted life years, QALYs),並以敏感度分析評估研究結果的穩健度。 研究結果:全膝關節置換術之醫院服務量閾值為120例/年,醫師服務量閾值為95例/年,全髖關節置換術之醫院服務量閾值為25例/年,醫師服務量閾值為10例/年,冠狀動脈繞道手術之醫院服務量閾值為70例/年,醫師服務量閾值為5例/年,頸動脈支架置入之醫院服務量閾值為55例/年,醫師服務量閾值為10例/年,二尖瓣置換與修復術之醫院服務量閾值為35例/年,醫師服務量閾值為15例/年。針對全膝關節置換術,病人於醫師服務量未達95例者接受手術,有較高之90日非計畫性再住院勝算;針對冠狀動脈繞道手術,病人於醫院服務量未達70例者接受手術,有較高的住院死亡與術後30日死亡勝算;針對頸動脈支架置入,病人於醫院服務量未達55例者接受手術,有較長之住院天數,病人於醫師服務量未達10例者接受手術,有較長之住院天數;針對二尖瓣置換與修復術,病人於醫師服務量未達15例者接受手術,有較高之住院費用、住院與出院後30日醫療費用以及住院與出院後90日醫療費用。針對全膝關節置換術、全髖關節置換術、冠狀動脈繞道手術、頸動脈支架置入以及二尖瓣置換與修復術,相較病人於低服務量醫院病人接受手術,病人於高服務量醫院接受手術具有成本效果。 結論:全膝關節置換術、全髖關節置換術、冠狀動脈繞道手術、頸動脈支架置入以及二尖瓣置換與修復術,存在醫院與醫師服務量閾值,病人於達到閾值之醫院與醫師接受手術,有較佳的照護結果,與較低之醫療利用。此外,病人於達到閾值之醫院接受手術具有成本效果。 Objectives: Previous studies have explored the relationship between volume and outcomes for total knee replacement (TKR), total hip replacement (THR), coronary artery bypass graft (CABG), carotid artery stenting (CAS), and mitral valve (MV) replacement and repair; however, little is known about whether there are optimal volume thresholds with regard to hospital and surgeon volume. Besides, few studies have adjusted the effects of potential unobserved confounders when examining the association between hospital and surgeon volume and outcomes. Moreover, it is unknown how cost-effective the implementation of hospital volume threshold is. There were three objectives in this study: (1) we determined the hospital and surgeon volume thresholds to achieve optimum outcomes; (2) we explored the association between hospital and surgeon volume threshold and outcomes; (3) we examined the cost-effectiveness of implemenetation of hospital volume threshold. Methods: With regard to the first and the second objectives, the database derived from Taiwan National Health Insurance Research Database and National Register of Deaths, which were provided by the Health and Welfare Data Science Center in Taiwan. As regards patients with TKR or THR, we included patients who received TKR or THR during January 1, 2015 and December 31, 2015. As regards patients with CABG, we included patients who received CABG during January 1, 2014 and December 31, 2015. As regards patients with CAS or MV replacement and repair, we included patients who received CAS or MV replacement and repair during January 1, 2011 and December 31, 2015. We applied restricted cubic spline, receiver operating characteristic (ROC) curve and Youden index to identify the optimal hospital and surgeon volume thresholds. Generalized estimating equations (GEE) and instrumental variables (IV) were used to discover the effects of hospital and surgeon volume thresholds on outcomes and healthcare utilization. With regard to the third objective, the data derived from the aforementioned database and related literature. The study population included that patients receiving TKR or THR in 2012, and patients receiving CABG, CAS or MV replacement and repair in 2013. Markov models were builded to assess the effects of implementation of the hospital volume thresholds on healthcare costs and quality-adjusted life years (QALYs). Sensitivity analysis was performed to evaluate the robustness of assessment. Results: As Regards TKR, the volume thresholds for hospitals and surgeon were 120 cases and 95 cases per year, respectively. As Regards THR, the volume thresholds for hospitals and surgeon were 25 cases and 10 cases per year, respectively. As Regards CABG, the volume thresholds for hospitals and surgeon were 70 cases and 5 cases per year, respectively. As Regards CAS, the volume thresholds for hospitals and surgeon were 55 cases and 10 cases per year, respectively. As Regards MV replacement and repair, the volume thresholds for hospitals and surgeon were 35 cases and 15 cases per year, respectively. Compared with patients who received TKR from surgeons with volumes reaching 95 cases a year, those received TKR from surgeons with volumes of fewer than 95 cases a year had higher odds of 90-day unplanned readmission. Compared with patients who received CABG from hospitals which reached 70 cases a year, those received CABG from hospitals which did not reach 70 cases a year had higher odds of in-hospital mortality and odds of 30-day mortality. Compared with patients who received CAS from hospitals with volumes reaching 55 cases a year and from surgeons with volumes reaching 10 cases a year, those received CAS from hospitals with volumes of fewer than 55 cases a year and from surgeons with volumes of fewer than 10 cases a year had higher length of stay. Compared with patients who received MV replacement and repair from surgeons who reached 15 cases a year, those received MV replacement and repair from surgeons who did not reach 15 cases a year had higher in-hospital costs, costs for the hospital stay and the first thirty days after discharge, and costs for the hospital stay and the first ninety days after discharge. With regatd to the cost-effectiveness assessments of five surgeries (TKR, THR, CABG, CAS, and MV replacement and repair), receiving surgeries from hospitals with volumes reaching the volume threshold was a cost-effective approach as compared with receiving surgeries from hospitals with volumes of fewer than the volume threshold. Conclusions: The hospital and surgeon volume thresholds can be identified for TKR, THR, CABG, CAS, and MV replacement and repair. Patients who received surgeries from hospitals with volumes reaching volume threshold and from surgeons with volumes reaching volume threshold have better outcomes and lower healthcare utilization. Moreover, receiving surgeries from hospitals which reach the volume threshold is a cost-effective approach. |
URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/7286 |
DOI: | 10.6342/NTU201901970 |
全文授權: | 同意授權(全球公開) |
電子全文公開日期: | 2024-08-26 |
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 | |
---|---|---|---|
ntu-108-1.pdf 此日期後於網路公開 2024-08-26 | 4.4 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。