手術量與醫療品質及利用之相關性探討 —以全膝關節置換術為例

Total knee arthroplasty is an expensive procedure compared with other medical procedure.

魏敏雄(Min-Hsiung Wei)

指導教授 ： 邱亨嘉

高雄醫學大學/健康科學院/公共衛生學研究所/碩士(2005年)

摘要

I
研究背景與目的：全膝人工關節置換術為健保費用支出的一大項目，文獻指出手
術後十年鬆脫的機率小於10%，15 年內鬆脫的機率約為15%（Rand JA. et
al,2003）。相關研究指出有高手術量的醫師，其豐富的手術經驗能夠達到好
的效果減少併發症，避免醫療資源浪費。本研究選擇利用率高且醫療費用高
的全膝人工關節置換術為例，來探討醫院、醫師手術量與醫療品質和服務利
1. 回溯性資料分析，利用全國健保申報次級資料，「住院醫療費用清單明細檔」
主處置ICD-9 CM procedure code=81.54 為研究對象，資料期間一年。
2. 控制病人主診斷與手術費用，排除骨惡性腫瘤、感染、骨折及單側膝關節置
換，必要排除1,648 筆、性別不詳7 筆，排除後，手術人次共7,806 筆。
3. 控制變項：病人、醫院特質；依變項為醫療品質指標包括在院死亡率和術後
3 個月內感染率和醫療服務利用指標包括平均住院天數及平均醫療費用。自
變項為醫院與醫師手術量，以四分位數法分為低、次低、次高、高手術量四
4. 以SPSS10.07 進行統計分析，描述樣本資料分佈、變項特性，以推論性統計
檢定手術量與醫療品質的相關性。
1. 醫院特質：在權屬別方面，私立醫院手術量為公立醫院的2.2 倍，私立醫院
在院死亡率低於公立醫院（adjusted OR=0.23、95%CI=0.07-0.78）。在層級
別方面，術後3 個月內感染率區域醫院比醫學中心高（adjusted OR=2.54、
95%CI=1.48-4.36），平均住院天數以醫學中心最短，平均醫療費用以地區醫
2. 病人特質：病人以女性佔多數（73.6%），平均年齡69.2 歲。65 歲以上病人
較易有術後3 個月內感染（adjusted OR=1.85、95%CI=1.30-2.64）及較長
的平均住院天數和較多的平均醫療費用。
3. 醫院手術量：高手術量組有較低的術後3 個月內感染率（adjusted OR=0.27、
95%CI=0.03-1.01），其醫療服務利用與手術量成反比。
4. 醫師手術量：術後3 個月內感染率和醫療服務利用與手術量成反比。

關鍵字

全膝人工關節置換術 ； 醫療品質 ； 醫療服務利用 ； 手術量

並列摘要

III
Total knee arthroplasty is an expensive procedure compared with other medical
procedure. According to previous reserches that hospital and/or physician had higher
volume of practice led to less complications and better outcome. The major purpose of
this study is to investigate the associations between provider volume and quality &
utilization with the example of primary total knee arthroplasty（TKA）.
This study design was a retrospective secondary data analysis. The sample
included a total number of 7,806 primary TKA cases (ICD-9-CM Code 81.54) who
received the operation by all National Health Insurance (NHI) contracted hospital from
October 2002 to September 2003. These NHI claim data sets was released from the
National Research Institutes. The computer software SPSS 10.07 for windows was used
to conduct descriptive and theorectical comparison and analysis. The total volume of
primary TKA procedures was divided into four groups by quartile method including
high, sub-high, sub-low and low volume. We examined the associations between the
annual volume of primary TKA performed in the hospitals and by the surgeons and the
indicators of quality ( in-hospital mortality rate and postoperative 3-month infection
rate ) and utilization (average in-hospital cost and length of stay in the hospital). The
analyses was adjusted for age, gender , comorbitidy and arthritis diagnosis. Moreover,
the quality and utilization differences in primary TKA will be compared by hospital
According to planned statistical analyses. The result indicated that：
1. The characteristics of hospitals：Provider volume in private hospitals was 2.2
times more than in public hospitals. In- hospital mortality rate in private hospitals
was less then in public hospitals（adjusted OR=0.23、95%CI=0.07-0.78）.
Postoperative 3-month infection rate was more than in regional hospitals than in
medical centers （adjusted OR=2.54、95%CI=1.48-4.36）.Average length of stay
in medical centers was the most shortest . Average medical cast in local hospitals
2. The characteristics of patients：The patients were mostly female（73.6%）with a
mean age of 69.2years.More than 65-year-old patient had higher postoperative
3-month infection rate（adjusted OR=1.85、95%CI=1.30-2.64）,longer average
length of stay and more average medical cost.
3. Provider volume in the hospital：Postoperative 3-month infection rate was
significantly lower in high volume group than in low group（adjusted OR=0.27、
95%CI=0.03-1.01）. Medical utilization was proportionally adverse.
4. Provider volume by the surgeon：Postoperative 3-month infection rate and
medical utilization indicator were proportionally adverse.

並列關鍵字

Provider volume. ； Total knee arthroplasty ； Medical utilization ； Medical quality

參考文獻

中文文獻

1. 簡麗年、朱慧凡、劉見祥、鍾國彪、曹昭懿、吳義勇、吳肖琪：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。台灣衛誌2003；

vol.22,No.2：118-126。

2. 吳敬堂、張香蘭、王文彥、侯毓昌：臨床路徑之介入對「論病例計酬」案件