不同材質人工髖關節對全人工髖關節置換術醫療利用之影響

Abstract

研究背景與目的：全人工髖關節置換術數目逐年攀升，所需醫療資源也隨之增加。過去研究多探討病患特質、醫師醫院特質對髖關節置換術醫療利用、術後結果的影響，然而人工髖關節材質也是影響因素之一。本研究即主要探討傳統材質及陶瓷材質間，病患使用不同材質人工髖關節的相關因素及其對醫療利用、術後結果的影響。 研究方法：本研究採橫斷式研究設計，分析全民健康保險資料庫2007-2008年承保抽樣百萬歸人檔。為追蹤出院後至90日內之醫療利用、術後品質，因此擷取2007年至2008年9月主手術碼為81.51：全人工髖關節置換術之病患為研究對象。以Wilcoxon Two-Sample Test及Kruskal-Wallis Test檢定醫療利用之差異，包含不同材質人工髖關節的使用情形、住院天數及手術期間、出院後30日、90日內醫療費用。為避免不同材質人工髖關節的使用上有選擇性偏誤，因此以羅吉斯迴歸計算病患使用陶瓷人工髖關節的傾向分數及其相關因素。複迴歸模型中則納入傾向分數校正可能的潛在效果，進一步分析使用不同材質人工髖關節對手術當次及術後短期內醫療利用之影響。術後結果的部份包含術後死亡、併發症及出院後再入院的情形。由於術後不良結果的發生數過少，因此無法進行變項檢定。 研究結果：395人次中，男性居多，45-54歲及65-74歲病患最多，手術主診斷最多為退化性關節炎、其次為股骨頭缺血性壞死，多數病患無任何併發症、需部份負擔。醫師年齡平均48.4歲、平均年資13年，醫院特質以財團法人醫院、台北分局、醫學中心的病患居多。在人工髖關節的利用上，使用陶瓷材質者較少，約14.7%。 所有病患平均住院天數為8天，手術當次花費114,995.0點，出院後30日、90日內醫療費用分別平均3,976.7點、7,253.4點。術後品質的部份，住院期間無病患死亡、出院後30日內及90日內死亡率分別為0.5%、0.8%；出院後30日內及90日內再入院率分別為2.3%、4.3%；住院期間及出院後30日、90日內併發症發生率分別為2.3%、6.6%、12.9%。描述性統計顯示，除了出院後再入院率以使用陶瓷材質者較高之外，其餘不良事件皆以使用傳統材質者為高。 經羅吉斯迴歸分析，65歲以上、地區醫院、北區、高屏分局的病患使用陶瓷人工髖關節的可能性顯著較低。而未納入傾向分數的逐步複迴歸模型中，使用陶瓷材質者當次手術醫療費用顯著較低，納入傾向分數後則與各依變項均未達顯著相關。 研究結論：納入傾向分數後，不同材質人工髖關節與手術當次、術後短期醫療利用皆未達顯著相關，但迴歸模型中仍可看出使用陶瓷材質者的醫療利用較低。建議後續研究者可運用縱貫性資料對長期醫療利用、術後結果進行分析、比較。

Background and objectives: Existing studies mostly looked at the effects of patient characteristics, surgeon and hospital characteristics on the utilization and outcomes of THR. However, types of bearing materials may also associate with medical utilization and outcomes of THR. Thus, the objectives of this study were to explore the medical utilization and outcomes of THR with different bearings and its related factors. Method: The design of this study is a cross-sectional study. We analyzed the claims data of the National Health Insurance from 2007 to 2008. In order to analyze the medical utilization and outcomes within ninety days after discharge, patients with ICD-9-CM procedure codes as 8151 between 2007 to September 2008 were selected. Wilcoxon Two-Sample Test and Kruskal-Wallis Test were used to analyze medical utilization of patients, including the usage of artificial hip joints, length of stay, hospitalization expenditure, and the medical fees within thirty and ninety days after discharge. The logistic regression was used to reduce the confounding effects by predicting the probability of using ceramic-on-ceramic bearings for each patient. The propensity score was included in multiple regressions to analyze the medical utilization of THR with different bearings and its related factors. Moreover, the study also reported the medical outcomes, which were evaluated by postoperative mortalities, readmission rates and complication rates. However, statistical tests can’t be performed due to few adverse medical events. Results: Among 395 cases, there were slightly more males, more patients aged 45-54 and 65-74. The most frequently seen primary diagnosis was osteoarthritis, followed by avascular necrosis. In addition, the average age of surgeons was 48.4 years old and the average seniority was 13 years. The majority of patients received health care from non-profit proprietary hospitals, medical centers, and in the jurisdiction of Taipei branch. In terms of the utilization of artificial hip joints, most patients adopted metal-on-polyethylene hip joints, only 14.7% use ceramic-on-ceramic bearings. The average length of stay was 8 days and the average hospitalization expenditure was 114,995.0 points. Additionally, the average medical fees within thirty and ninety days after discharge were 3,976.7 and 7,253.4 points, respectively. In terms of medical outcomes, the mortalities within index hospitalization, thirty days and ninety days after discharge were 0.0%, 0.5% and 0.8%, respectively. The readmission rates within thirty days and ninety days after discharge were 2.3% and 4.3%, respectively. The complication rates within index hospitalization, thirty days and ninety days after discharge were 2.3%, 6.6% and 12.9%, respectively. All the adverse medical events occurred more frequently in the patients who adopted metal-on-polyethylene hip joints except for readmission rates. As shown by the logistic regression, patients who aged 65 years and older or received health care from district hospitals, Northern region and Kao-Ping branches were less likely to adopt ceramic-on-ceramic bearings. Furthermore, the hospitalization charge was significantly lower for patients with ceramic-on-ceramic bearings in the stepwise multiple regressions without propensity score. But it became statistically insignificant in the models with propensity score. Conclusion: The multiple regressions with propensity score indicated that THR with different bearings was not significantly related to hospitalization expenditure and medical utilization after discharge. Nevertheless, the results still displayed that THR with ceramic-on-ceramic bearings was associated with lower medical utilization. We suggest that further research can conduct longitudinal studies to explore and compare long-term medical utilization and outcomes.