健保局DRGs對台灣病例變異解釋力之初探

A Pilot Assessment of the Variation and Explanation of the BNHI-DRGs for Cases in Taiwan

10.6288/TJPH2005-24-06-09

陳琇玲(Hsiu-Ling Chen)；楊銘欽(Ming-Chin Yang)；薛亞聖(Ya-Seng Hsueh)；楊捷如(Chiech-Ju Yang)

DRGs ； 變異係數 ； 解釋能力 ； 健康保險 ； DRGs ； coefficient of variation ； explanation ； health insurance

台灣公共衛生雜誌

24卷6期（2005 / 12 / 01）

548 - 560

繁體中文

Objectives: The purpose of this study is to determine the coefficient of variation (CV) and the explanation of the BNHI-DRGs in order to provide suggestions for Taiwan. Methods: The research data is derived from the 2001 National Health Insurance Research Database of the National Health Research Institutes. After blending two data files, combining the same patient data, and also dropping out primary care datasets, we used the BNHI-DRGs (version 1) for group cases. Then we analyzed the coefficient of variation (CV) and explanation (R^2) after deleting the outliers. Results: We determined the CV and R^2 of the charges and lengths of stay (LOS) for 495 DRGs and R^2 for each MDC. For both the charges and LOS, most CV values of the DRG are between 0.5 and 1.0. The overall charge explanation value is 0.39 and LOS explanation is 0.28. After excluding the case payment corresponding DRGs, the charge explanation is found to be 0.35 and LOS explanation 0.27, and there are significant negative correlations between CV and R^2 (p＜0.01) with r=-0.5 for both the charge and LOS. Conclusions: (1) Whether or not the case payment corresponding DRGs are included, the BNHI-DRGs provides a better explanation for 2001 NHI inpatient charges than studies in other countries, indicating that it may be appropriate for DRGs to be put under a new payment system in Taiwan. (2) The DRGs amendment according to the CV of each DRG is a correct direction. However, referring to the explanation of individual DRGs is strongly suggested. (3) The BNHI should initially research DRGs with the explanation being near zero, medical partitioning, MDC4, MDC7, and MDC11. (4) The charge explanations for neonates (MDC15) in Taiwan are higher than the research results in other countries, but still less then the cases overall. When increasing neonatal DRGs, the BNHI should consider the possibilities of international comparisons occurring in the future. (5) The BNHI should set the criteria for identifying additional DRGs where a CC (complication and comorbidity) split appears most justified. Additional DRGs should not be split based solely on the presence or absence of a CC because hospitals may respond by changing coding practices to increase total payments, which would not represent a real increase in the severity of the overall mix of cases.

Objectives: The purpose of this study is to determine the coefficient of variation (CV) and the explanation of the BNHI-DRGs in order to provide suggestions for Taiwan. Methods: The research data is derived from the 2001 National Health Insurance Research Database of the National Health Research Institutes. After blending two data files, combining the same patient data, and also dropping out primary care datasets, we used the BNHI-DRGs (version 1) for group cases. Then we analyzed the coefficient of variation (CV) and explanation (R^2) after deleting the outliers. Results: We determined the CV and R^2 of the charges and lengths of stay (LOS) for 495 DRGs and R^2 for each MDC. For both the charges and LOS, most CV values of the DRG are between 0.5 and 1.0. The overall charge explanation value is 0.39 and LOS explanation is 0.28. After excluding the case payment corresponding DRGs, the charge explanation is found to be 0.35 and LOS explanation 0.27, and there are significant negative correlations between CV and R^2 (p＜0.01) with r=-0.5 for both the charge and LOS. Conclusions: (1) Whether or not the case payment corresponding DRGs are included, the BNHI-DRGs provides a better explanation for 2001 NHI inpatient charges than studies in other countries, indicating that it may be appropriate for DRGs to be put under a new payment system in Taiwan. (2) The DRGs amendment according to the CV of each DRG is a correct direction. However, referring to the explanation of individual DRGs is strongly suggested. (3) The BNHI should initially research DRGs with the explanation being near zero, medical partitioning, MDC4, MDC7, and MDC11. (4) The charge explanations for neonates (MDC15) in Taiwan are higher than the research results in other countries, but still less then the cases overall. When increasing neonatal DRGs, the BNHI should consider the possibilities of international comparisons occurring in the future. (5) The BNHI should set the criteria for identifying additional DRGs where a CC (complication and comorbidity) split appears most justified. Additional DRGs should not be split based solely on the presence or absence of a CC because hospitals may respond by changing coding practices to increase total payments, which would not represent a real increase in the severity of the overall mix of cases.

1. 3M(2000).Diagnosis Related Groups Definitions Manual Version 18.0. 1st ed.3M,595-602|795-9.
2. Calore KA,Iezzoni L(1987).Disease staging and PMCs-Can they improve DRGs.Med Care,25,724-37.
3. Centers for Medicare & Medicaid Services (CMS)(2003).Medicare Program; Changes to the hospital inpatient prospective payment systems and fiscal year 2004 rates; proposed rules.Fed Regist,68,27153-422.
4. Centers for Medicare & Medicaid Services (CMS)(2003).Medicare Program; Changes to the hospital inpatient prospective payment systems and fiscal year 2004 rates; final rule.Fed Regist,68,45345-672.
5. Centers for Medicare & Medicaid Services(CMS)(2003).Medicare Program; Change in methodology for determining payment for extraordinarily high-cost cases (cost outliers) under the acute care hospital inpatient and long-term care hospital prospective payment systems.Final rule|Fed Regist,68,34493-506.
6. Forbes JF,Pickering RM(1988).Development of a neonatal case-mix classification.Med Care,26,1033-45.
7. Freeman JL,Fetter RB,Park H(1995).Diagnosis-Related Group refinement with diagnosis-and procedure-specific comorbidities and complications.Med Care,33,806-27.
8. Kawabuchi K(2000).Payment systems and considerations of case mix-Are diagnosis-related groups applicable in Japan.Pharmacoeconomics,18,95-110.
9. Lichtig LK,Knauf RA,Bartoletti A(1989).Revising diagnosis-related groups for neonates.Pediatrics,84,49-61.
10. Mccrone P(1995).Predicting mental health service use: diagnosis based systems and alternatives.J Ment Health Adm,4,31-40.
11. McGinnis GE,Osberg JS,DeJong G,Seward MI,Branch LG(1987).Predicting charges for inpatient medical rehabilitation using severity, DRG, age, and function.Am J Public Health,77,826-9.
12. Muldoon JH(1999).Structure and performance of different DRG classification systems for neonatal medicine.Pediatrics,103,302-18.
13. Phibbs CS,Phibbs RH,Pomerance JJ,Williams RL(1986).Alternative to diagnosis-related groups for newborn intensive care.Pediatrics,78,829-36.
14. Resnick MB,Ariet M,Carter RL(1986).Prospective pricing system for tertiary neonatal intensive care.Pediatrics,78,820-8.
15. Summary of current status of All Patient Refined Diagnosis Related Groups(APR-DRGs)
16. 中央健康保險局(2004)。全民健康保險住院診斷關聯群支付方案規劃報告
17. 第一、二版DRG項目對照表(499DRG vs 976 DRG)
18. 中央健康保險局(2002)。住院病例組合架構說明。總額制度下Case Mix醫療費用支付及審查之應用，4-1-9。
19. 黃瑞美｜鄭茉莉｜林富滿｜林秀美｜陳琇玲｜楊捷如｜李宣緯(2003)。建立醫院病例組合指標－以新生兒疾病與疾患為例。行政院衛生署九十一年度委託研究計畫(DOH91-NH-1018)
20. 黃肇明(2002)。住院病例組合支付制度方式下之審查規劃。總額制度下Case Mix醫療費用支付及審查之應用，1-1-13。

1. 錢才瑋、王文中(2007)。健保實施DRGs對醫學中心之影響：CTT與IRT之比較。致遠管理學院學報，2(2)，89-106。
2. 楊雅淑、黃偉堯(2013)。高危險妊娠因素與出生體重對新生兒住院醫療費用的影響。臺灣公共衛生雜誌，32(3)，240-250。