全人工髖關節及膝關節置換術服務量與結果之關連性探討

Ａssociation between hospital and surgeon volume with outcome in total hip and total knee replacement

羅賢益(Hsien-Yi Lo)

指導教授 ： 蘇喜

國立臺灣大學/公共衛生學院/醫療機構管理研究所/碩士(2005年)

https://doi.org/10.6342/NTU.2005.00632

摘要

1970年代起，關於照護品質與醫師及醫院服務量間相關探討議題，在國外已廣泛討論，在國內照護結果、醫療費用與醫院及醫師服務量間相關研究則為數有限。一般文獻中的研究以癌症及高難度的手術為主，少數幾篇有關於全人工關節置換術的報告。有鑒於人口老化與保險給付造成全人工關節置換術的利用率逐年增加，耗用資源相當大。好的照護結果可以使得醫療利用達到最佳效率，因此希望以實證資料來分析醫院與醫師服務量與照護結果的相關性。
本研究利用1999年1月至2001年12月全民健保資料庫，以全人工髖關節置換術及全人工膝關節置換術為對象，分析醫院服務量、醫師服務量與照護結果(死亡、併發症)的關聯性，並依病人特性、醫院特性，研究其與照護結果的相關性。追蹤期間為當次手術住院、手術後30天、手術後90天、手術後一年。結論是不管追蹤期間，均可以得到一致的結果及趨勢，亦即高服務量與死亡率及併發症發生率呈負向相關。追蹤期間愈長，得到的趨勢愈明顯。這不僅見於以四十、六十、八十百分位做分組，也見於我們依服務量斜率所做新的服務量分組。若研究實際醫院或醫師服務量與照護結果的相關性，也發覺較高死亡率及併發症發生率較多見於低服務量的醫院及醫師。
同一次住院施行兩側關節手術(髖或膝)，比不同次住院施行兩側關節手術及在研究期間只施行一側手術的結果為差，不過沒有統計上意義。
全人工髖關節置換術的診斷包括退化性關節炎、缺血性壞死、類風濕性關節炎，前兩者所佔比例相當。全人工膝關節置換術以退化性關節炎為主，不過診斷對於結果影響不大。男性患者、年齡愈大者在全人工髖關節置換術及全人工膝關節置換術的死亡率及併發症發生率較高。
大約有四分之三的病人，接受高服務量醫院或醫師進行手術，已有自然的集中化效果，所以對於區域化(regionalization)的需求並不迫切。倒是如何在專科訓練過程中加強手術實際操作，以累積其經驗，避免獨當一面時，因經驗不足造成死亡率或併發症發生率的居高不下。
另外以邏輯式廻歸，可以得到在全人工髖關節置換術中，性別、年齡、醫師服務量有意義。在全人工膝關節置換術中，除了性別、年齡、醫師服務量有意義外。醫院層級、醫院權屬別、醫院服務量也有意義。
因此，在全人工髖關節及膝關節置換術的關聯性探討可以得知，高服務量醫院或醫師所造成的死亡率及併發症發生率較低，與文獻報告大致一樣。

關鍵字

全人工髖關節置換術 ； 全人工膝關節置換術 ； 醫院服務量 ； 醫師服務量 ； 照護結果

並列摘要

The relationship between volume and outcome has been investigated since 1970. Most of the literatures focused on the cancer surgery and highly technique- demanded surgery, and few of those talked about joint replacement. Due to aging of population and increasing payment for the joint replacement, a good outcome of the health care would achieve the effective utilization of the resources. An empirical evidence of the association between the hospital and surgeon volume and outcome in total joint replacement could give us some answers.
This study examined the relationship between the volume of total hip replacements and total knee replacements by hospitals and surgeons and the postoperative rates of death and complications (revision, dislocation, deep infection, periarticular fracture). A nation wide hospital discharge registry was used to identify patients who had an elective total hip replacement and total knee replacement between 1999 and 2001,and the follow up period included index hospitalization, 30 days postoperatively, 90 days postoperatively, 1 year postoperatively. The number of hip replacement and knee replacement performed during the study period was determined by each provider. The providers ware divided into four groups on the basis of fortieth, sixtieth and eightieth percentiles, and another way according to the slope change of the volume curve was divided into low, medium, and high volume. The real distributions of the provider volume and death or complication rate were also shown in this study.
Patients managed by the low-volume providers tended to have higher mortality rates, more complication rates during the index hospitalization and further follow up period.
Single joint replacement, bilateral joints replacement simultaneously, or separate operation during different hospitalization were compared using this large scale data. Although it had some difference,there was no static significantly.
About three –fourths patients received operation by the high volume provider, so it seemed to have natural regionalization tendency in this study. Enhancement of the training program to improve the experience of the practitioner, so that to get better outcome would be more important.

並列關鍵字

total hip replacement ； total knee replacement ； hospital volume ； surgeon volume ； outcome

參考文獻

48.簡麗年、朱慧凡、劉見祥、鍾國彪、曹昭懿、吳義勇、吳肖琪 (2003)。醫院、醫師手術量與醫療品質之關聯性探討－以全股(髖)關節置換為例。臺灣公共衛生雜誌，22(2)，118-126。

64 魏榮南等：1996_2000年台灣帝屈糖尿病盛行率與住院率，台灣衛誌 2002： 21(3) 173-80.

1. Adams DF, Fraser DB, Abrams HL:The complications of coronary arteriography. Circulation 48:609, 1973.

2. Luft HS, Bunker JP, Enthoven AC. Should operations be regionalized? The empirical relation between surgical volume and mortality. N Engl J Med. 1979;301:1364-9.

3. Luft HS:The relation between surgical volume and mortality: an exploration of causal factors and alternative models.Med Care 18:940, 1980.