|摘要: ||中央健康保險局為了能落實財務分擔風險，從民國91年7月實施醫院總額支付制度，以醫院自主管理模式，不論各階段都以「十四日內再住院率」作為監測醫療品質的替代指標。「再住院率」同時也是台灣品質指標計劃（Taiwan Quality Indicator Project，TQIP）重要的醫療品質監測指標之一，醫院評鑑基準亦將「十四日內再住院」列為醫療品質改善活動檢討的項目，未來實施前瞻性診斷關聯群支付制度（DRGs /PPS），醫療機構為成本控制考量，部分病患可能在疾病未痊癒之下提早出院、醫療提供者取巧分段住院或其他相關因素影響病患再住院，有關再住院管理仍是健保局持續加強審查的重點，亦是醫院管理重要主題之ㄧ。
本研究目的為分析醫院再住院率、確認出院後90日內不同間隔日數再住院風險影響因素，進而建構再住院病患之預測模式。本研究從某醫學中心94年1月1日至94年12月31日期間，選用至少住院一次病患，總計5,931人為研究對象，得再住院群組樣本3,026人、對照組為非再度住院2,905人，以該研究對象的疾病分類資料檔內容作為分析，運用SPSS 10.0 for Windows 套裝軟體系統進行描述性及推論性統計分析。
To realize the financial risk reduction, the Bureau of National Health Insurance has implemented the Global Budget Payment System by the self management mode and considered the rate of readmission within 14 days for all stages a substitute indicator for medical quality monitoring since July, 2002. Besides, the remarked rate is an essential indicator for the above monitoring by Taiwan Quality Indicator Project (TQIP). Moreover, attributed to the mentioned readmission listed in the hospital accreditation standard, handling readmission becomes an important issue of hospital management. According to scholars, due to Diagnostic Related Groups/ Prospective Payment System (DRGs/PPS) -caused hospital cost control, some uncured patients may be discharged early, and medical providers may be tricky for segmented admission, or other associated factors may affect readmission. Therefore, despite Taiwan-developed diagnosis related groups systems (Tw-DRGs) by the rate executed in 2008, the rate remains intensively examined by the Bureau.
The purpose of this study is to analyze the readmission rates, identify the risk factors of readmission within 90 days, and establish the predictive pattern. A total of 5,931 recipients with at least one hospitalization were selected by a medical center from January 1st, 2005, to December 31st, 2005 for this study, including 3,026 recipients who had hospital readmissions and 2,905 hospitalized but non-readmitted recipients as a control group. The International Classification of Diseases database-based data of the two groups were descriptively and inferentially analyzed by the SPSS 10.0 for Windows software package.
Resultantly, the rate of readmission within 90 days, 14 days, 15-30 days, 31-60 days, and 61-90 days accordingly was 17.84%, 6.31%, 5.96%, 3.60%, and 1.97%. The average cost for the control group and the fist readmission was ＄50,748 and＄68,347, respectively. Concerning the high medical cost (＄60,001 more), the percentage of the readmission group (33.5%) was 11.6% more than that of the control group (21.9%). About the lower medical cost (＄40,000 less), the percentage of the latter was 62.4%; that of the former, 49.4%.
The logistic regression-analyzed risk factors of readmission within 90 days with different intervals were the senile, longer hospital days, fee-for-service cases, cases of more diseases, and ischemic heart disease. Additionally, the risk factors of readmission within 14 days, 15-30 days, 31-60 days, and 61-90 days orderly were significantly associated: (1) male, doctors with less than 5 years working experience, other disorders of urethra and urinary tract, cholelithiasis, malignant neoplasm of liver and intrahepatic bile ducts, and heart failure; (2) doctors with less than 5 years working experience, cholelithiasis; (3) cholelithiasis, malignant neoplasm of liver and intrahepatic bile ducts, and heart failure; (4) malignant neoplasm of liver and intrahepatic bile ducts.
The efficient application of the readmission predictive pattern may detect few of the patients at high-risk of readmission. The high risk factors, especially for the senile patients, should be avoided by early discharge planning and medical care process strengthening. For prospective Tw-DRGs fulfillment, scaling-up of medical organization-practiced clinical path which enforces medical quality, prevents readmissions and controls hospital costs is recommended.
Keyword: Readmission, Readmission rate, Quality of medical care,Outcome indicators