中文摘要
目的：探討健保住院安寧支付制度下醫院收益之影響因素。方法：以2004年至2006年全民健保安寧住院病患為研究對象，將病患醫療費用與安寧申報費用之間差異定義為安寧收益。利用階層線性迴歸模式探討影響安寧收益之相關因素。結果：安寧收益為正值的病患，於2006年有90.6％比2004年增加2.4％。醫院平均安寧收益超過20,000點之醫院，2004年占50％至2006年增加為76.7％。病患安寧收益受醫院評鑑等級、病患住院日與醫院平均住院日聯合效果以及病患癌症別顯著地影響。醫學中心收益少於地區醫院，病患安寧收益隨著住院天數增加而增加，醫院平均住院日會減弱病患住院日對安寧收益的影響；口腔癌相較於其他癌症安寧收益較少。 結論：現行住院安寧療護論日計酬制，應考量不同醫院評鑑等級之醫療費用差異，至於醫院整體平均住院日對於病患住院日與安寧收益之負向調節效果，值得繼續作更深入討論。 Abstract
Objectives: No empirical study to date has investigated the impact of Taiwan National Health Insurance ﹙NHI﹚ Pilot Project pricing for Inpatient Hospice Services and its relations with patient and hospital characteristics. This study employed a hierarchical linear modeling ﹙HLM﹚ to explore the importance of both patient-level and hospital-level attributes in determining income benefit of Inpatient Hospice Services under the Taiwan NHI Pilot Project pricing system. Methods: The hospice admission of the administrative claims data of beneficiaries in Taiwan between 2004 and 2006 were used. The income benefits for Inpatient Hospice Service ﹙income benefits﹚were measured as the differences between medical expenses and hospice benefits. Data analysis was undertaken using descriptive statistics and HLM. Results: We found that 90.6 percent hospice services with a positive income benefit in 2006, compared to the previous year, 2004's figure was 2.4 percent lower. As to hospital’s mean income benefit more than 20,000 RVU ﹙relative value unit﹚, there were 50 percent hospitals in calendar year 2004 and up to 76.7 percent in 2006. Results of HLM demonstrated that hospitals’ income benefits were different across hospitals significantly. The income benefits were associated with characteristics on both the patient-level including length of stay, type of cancer and the hospital-level including accreditation level, the interaction of hospital’s mean length of stay and patient’s length of stay. The income benefits were positively related to length of stay; whereas the patients’ length of stay aggregated the hospital level diminished the relationship between income benefit and length of stay. The models also indicated that the income benefits of medical centers were lower than community hospitals when inpatient hospice services were priced equally among overall hospitals in 2006. Conclusion: Empirical results show that length of stay is associated with income benefits and its relations are confounded by hospitals’ aggregated length of stay. The income benefits are significant differences among types of cancer and across hospitals. Implications for research in homogeneous Taiwan NHI Pilot Project pricing system are considered.
Key Words: hospice care; hospice income benefit; length of stay; per-diem payment
