癌症病患初始階段醫療資源耗用與死亡率分佈趨勢
及影響因素之探討-以肺癌、肝癌、結直腸癌、乳癌、胃癌為例

研究目的針對國人罹患癌症分佈趨勢及健保財務負擔的現況，進而探討國人常見癌症病患在確診前六個月至確診後一年期間(初始階段)為醫療成本花費的高峰期，因此探討不同癌症的醫療資源及成本效益評估，進而提出本研究之目的。一、探討五種癌症病患，初始階段(initial phase)之醫療資源耗用與死亡率分佈趨勢二、探討五種癌症治療模式，初始階段醫療資源耗用與死亡率之差異性三、探討影響五種癌症病患，初始階段醫療資源耗用與死亡率之相關因素四、比較各國五種癌症病患，初始階段醫療資源耗用與死亡率之差異性分析研究方法一、本研究採回溯性縱貫性研究設計，利用國家衛生研究院釋出之「全民健康保險研究資料庫」-「20萬人承保抽樣歸人檔」之次級資料建立、將1996至2007年主/次診斷為肺癌、肝癌、結直腸癌、女性乳癌、胃癌的患者為研究樣本共6314人，探討病患確診當次、確診前半年、確診後一年接受治療之醫療資源耗用(住院天數、醫療總費用、手術治療費用、化學治療費用、放射治療費用、荷爾蒙治療費用)並計算初始階段，就醫花費與死亡率的分佈趨勢及影響因素。二、採用SPSS 14.0 版本之統計套裝軟體工具進行資料整理與統計分析，主要的統計方法包括：描述性統計（次數分配、平均數、標準差及百分比等）與卡方檢定(Chi-square test)、t檢定、one-way ANOVA檢定、Effect size(改變幅度)、Bootstrapping變異法與複迴歸分析(Multiple Regression Analysis)、Cox存活分析等推論性統計。研究結果一、五種癌症病患：肺癌(1451例)、肝癌(1691例)、結直腸癌(1833例)、女性乳癌(835例)、胃癌(504例)，以結直腸癌患者居多。二、在1996至2007年，肺癌、肝癌、結直腸癌患者初始階段之住院天數有隨時間減少之趨勢；在三個年代(1996-1999、2000-2003、2004-2007)、確診當次之住院天數，五種癌症皆隨時間而減少之趨勢；確診前半年之住院天數，肝癌、結直腸癌亦為減少之趨勢；確診後一年之住院天數，除肺癌與肝癌隨時間減少外，結直腸癌、乳癌、胃癌則為增加之趨勢。三、初始階段醫療總費用方面，五種癌症病患在1996至2007年分佈趨勢皆有顯著增加；在三個年代、三個區間(確診前半年、確診當次、確診後一年)，除肺癌、肝癌、結直腸癌、乳癌病患之確診當次醫療總費用有減少之情況外，五種癌症病患的初始階段醫療總費用整體而言亦為增加。四、五種癌症病患之死亡率隨時間有顯著減少之趨勢。五、若有施行手術/化學治療/放射治療/荷爾蒙治療，初始階段之住院天數與醫療總費用比未施行四種治療模式患者皆有顯著增加。然而，有施行手術/化學治療之死亡率比未施行手術/化學治療患者有下降之現象，唯肝癌和乳癌病患有顯著下降。六、複迴歸分析顯示人口學特性、臨床特性、醫院特性皆會顯著影響初始階段住院天數、手術費用、三種輔助性療法費用與醫療總費用。七、存活分析顯示影響癌症病患存活狀況之顯著因子為年份、CCI、醫師服務量、癌症類別、有無施行手術治療/放射治療。結論與建議研究結果發現五種癌症病患之人口學特性、臨床特性、醫院特性與住院天數及費用間皆有顯著的相關。整體而言，初始階段住院天數隨時間有顯著減少的趨勢，而醫療總費用隨時間有顯著增加的趨勢，死亡率隨著時間顯著下降的趨勢，顯示醫療科技的進步，醫療水準的提升，健保政策制度的功效，提升了醫療品質，然而醫療費用卻不斷的成長，如何控制逐年高漲的健保費用及資源分配是衛生當局現今重要課題。經由本研究五個癌症之研究數據，了解不同癌症疾病資源耗用分佈趨勢，將全面的了解不同癌症的經濟負擔，或許可以提供衛生主管機關對於規劃未來醫療預算以及資源分配擬定之參考，建立更完善的衛生保健政策，維護全民的健康，增進民眾的福祉。

關鍵字

癌症；醫療資源耗用；住院天數；死亡率；化學治療；放射治療

並列摘要

Purpose The purpose of study is to analyze distribution of cancer patients and to discuss the highest cost at 6 months before diagnosis and 1 year after diagnosis. Therefore, the study evaluates the difference between medical resource utilization and cost-effective on five common cancers. Methods This retrospective study used National Health Insurance Research Database and 6,314 patients whose main diagnosis and secondary diagnosis were lung cancer, liver cancer, colorectal cancer, breast cancer, and stomach cancer were included into the study. Furthermore, this study also evaluated medical resource utilization (length of days, medical cost, and cost of surgery, chemotherapy, radiotherapy, and hormone therapy ) at the initial cancer care, 6 months before diagnosis, and 12 months after diagnosis and this study finally conducted trend analysis of medical resource utilization and mortality. Additionally, this study developed descriptive statistics (such as: distribution of times, means, SD, percentage, and so on) and inferential statistics (such as: t-test , one-way ANOVA, effect size, bootstrapping, multiple regression analysis, Cox survival analysis, and so on). Results 1. 5 cancer patients: The results showed lung cancer (n=1451), liver cancer (n=1691), colorectal cancer (n=1833), breast cancer (n=835), and stomach cancer (n=504) in the study and the most of patients were colorectal cancer. 2. The result also showed length of days of lung cancer, liver cancer, and colorectal cancer patients became decreased in the initial phase during 1996 and 2007; length of days at the initial cancer care of 5 cancer patients became decreased; length of days of liver cancer and colorectal cancer patients became decreased at 6 months before diagnosis; length of days of colorectal cancer, breast cancer, and stomach cancer patients became increased at 1 year after diagnosis. 3. In the initial phase, medical cost of 5 cancer patients significantly increased during 1996 and 2007. 4. The result showed mortality of 5 cancer patients became decreased. 5. In the initial phase, the patients who accepted surgery, chemotherapy, radiotherapy, and hormone therapy had significant increased. However, mortality was reduced when patients accepted surgery and chemotherapy among liver cancer and breast cancer patients. 6.The multiple regression analysis showed demographic characteristic, clinical characteristic, and hospital characteristic were significant affect length of days, surgery, cost of adjuvant therapy, and medical cost in the initial phase. 7. The survival analysis showed CCI, physician volume, types of cancer, and surgical treatment/ radiotherapy were the important impact factors. Conclusion The results showed demographic characteristic, clinical characteristic, and hospital characteristic were significantly related to length of days and medical cost. Generally, both length of days and mortality decreased gradually but medical cost increased gradually. The study also showed medical technology and quality of medicine has improved, and the medical cost increased. Therefore, how to allocate the scarce utilization allocation is the most important issue for health institutions. According to the results, we understand trends of medical resource utilization and economic burden on different cancers. Careful management of these factors can enhance efficiency in allocating scarce hospital resources. Additionally, treatment strategies adopted by high-volume hospitals and surgeons should be further analyzed and utilized more widely.