全民健保乳癌試辦計畫對乳癌手術醫療利用之影響-以乳房保留手術
與放射治療為例
Effect of the NHI Breast Cancer Quality Enhancement Pilot Project on Breast Cancer Surgery Utilization Results from the Breast
Conserving Surgery and Radiation
中文摘要
中央健康保險局(以下簡稱健保局)於 2001 年 10 月對於乳癌支付制度有一個重大 改變，即試辦以乳癌疾病療程為支付基準的論質計酬支付方式。本研究目的在於 探討乳癌試辦計畫對施行乳房保留手術機率、乳房保留手術後施行放射線治療機 率及乳癌手術後一年內醫療費用之影響。本研究之樣本為 2001 年至 2004 年確診 為乳癌且於確診日後一年內施行乳癌手術之患者計 15,574 人。本研究採用類實 驗設計的研究方法進行分析。其中實驗組醫院有 6 家，共計病患 4,306 人(試辦前 2,103 人，試辦後 2,203 人)；對照組醫院共 87 家，有 11,268 病患。羅吉斯迴歸 分析結果發現未有足夠證據顯示乳癌試辦計畫之實施與乳房保留手術機率之增 加有顯著關連，其主要原因為對照組施行乳房保留手術之比例也同時大幅成長。
在乳房保留手術患者是否施行放射線治療部分，樣本數 3,253 人，其中實驗組有 968 人(試辦前 397 人，試辦後 571 人)，對照組有 2,285 人。羅吉斯迴歸分析顯 示實驗組試辦前相較對照組其乳房保留手術後施行放射線治療機率之勝算比為 4.56，而實驗組試辦後相較對照組之勝算比更高達 38.04。
至於乳癌患者手術後一年內醫療利用部分，15,574 患者中具有完整健保申報資料 者共計 15,557 人做為研究對象。本研究以複迴歸分析發現實驗組試辦前與對照 組之醫療費用並無統計上的顯著差異，但是實驗組試辦後之醫療費用卻低於對照 組。
由本研究結果及乳癌試辦計畫的複雜性與較低的醫院參與率，本研究之政策建議 如下：首先，健保局應全面檢討乳癌試辦計畫之實施成效，尤其應思考乳癌試辦 計畫定位是朝向獎勵品質逐年提升的醫院或獎勵好品質的醫院；其次，論質計酬 的精神在於鼓勵醫療品質而非特定治療方法，因此乳房保留手術支付點數之訂定 宜回歸其技術困難度，而財務誘因則宜與品質監控指標結合，以獎勵乳房保留手 術品質卓著醫院；第三，乳癌試辦計畫醫療費用之申報方式應與其他一般醫療費 用一致，以減少行政成本；最後，本研究建議健保局以申報資料結合品質監控指 標定期檢討調整總療程之支付點數。
英文摘要
The Bureau of National Health Insurance (NHI) started to experiment an
episode-based pay-for-performance program for breast cancer, the NHI Breast Cancer
Quality Enhancement Pilot Project, in October of 2001. This study aims to examine whether or not the pilot project had impacts on the increase of the rates of
breast-conserving surgery and subsequent radiation therapy after breast-conserving surgery, and on the one-year utilization of medical care resources for breast cancer patients.
The sample consisted of 15,574 patients who had been confirmed with breast cancer between 2001 and 2004 and undergone breast cancer surgeries within one year after confirmation. In the quasi-experimental design, the experimental group (6 hospitals) had 4,306 patients (2,103 and 2,203 in the pre- and post-pilot-project groups,
respectively), while the control group (87 hospitals) had 11,268 patients. Results from the logit model found no sufficient evidence to indicate the pilot project had increased the rate of breast-conserving surgeries for patients in the experimental group, mainly because the breast-conserving surgery rates were also increasing in the control group.
For those who underwent breast-conserving surgeries (3,253 patients), 968 were in the experimental group (397 and 571 in the pre- and post-pilot-project groups). The logit model confirmed that patients in the pre-pilot-project group were more likely to receive radiation therapies than those in the control group, with an odds-ratio 4.56 approximately, whereas patients in the post-pilot-project group had even higher likelihood to undergo radiation therapies than those in the control group, and the odds-ratio was about 38.04.
Regarding the one-year utilization of medical care resources after and including breast cancer surgery, 15,557 patients had available data among those 15,574 cases. The multiple regression model demonstrated that the medical expenses of patients in the pre-pilot-project group were not statistically different from those in the control group, but patients in the post-pilot-project group had lower medical expenses than those in the control group.
Given the study results, and the complicated design and low participations of the pilot project, the study suggests that the Bureau of NHI may consider the following for policy change. First, who should be rewarded, those who are already outstanding or those who strive to improve their quality over time? How effective the current program has been? Second, the payment incentive in a pay-for-performance program should reward good quality rather than particular procedure. Therefore, the current higher payment for breast-conserving surgery should be revised so that the payments for different surgeries are all based on their costs. Third, the special claim reporting rule for the pilot project should be simplified to decrease administrative costs and encourage hospital participation. Finally, given the availability of the utilization data, the payment rates should be reviewed and revised periodically.
