Impact of the Global Budget Payment System of Taiwan’s NHI on Outpatient Visits and Expenditures, 2001-2005

Presenter: Ming-Chin Yang, College of Public Health, NTU

Abstract

Purpose: Taiwan’s health insurance (NHI) implemented the Hospital Global Budget Payment System (HGBPS) in 2003. One of the system's purposes is to use financial incentives to encourage hospitals to exercise self-regulation and to reduce the proportion of cases that can be treated primarily in the clinics. The purpose of this study was to use the claims data of the Bureau of NHI from 2001 to 2005 to examine whether there was any change in the proportion of disease classification before and after the implementation of the HGBPS.

Methods: This study was a secondary data analysis. In terms of disease classification, according to a previous study by one of our authors, diseases could be classified into three types (ABC classification). For example, those can be treated primarily in clinics are categorized as type A diseases; while those can only be treated in larger hospitals are categorized as type C diseases; the remaining diseases can be seen at any level of healthcare institutions are type B diseases. In this study we first classified each disease into one of the ABC categories. Then we used monthly claims data of each facility as the unit of analysis and applied the generalized estimation equation (GEE) model to analyze the data.

Results and Discussion: After controlling for confounding factors, the overall outpatient visits of hospitals or clinics decreased slightly after the implementation of either primary care global budget payment system (PCGBPS) or HGBPS. As for the expenditure claimed, the change was -4.96% (Z=-14.90, p<.001) after implementing the PCGBPS, but it was not significant (Z=0.39, p>.05) after HGBPS. However, we didn’t observe more type A diseases shifted from hospitals to clinics as expected after the HGBPS in this study.

Conclusion: Hospitals did not reduce the proportion of type A cases significalty after the HGBPS. Future study should consider adopting more sensitive indicators or specific disease categories to examine whether patients shifted between clinics and hospitals after implementing the HGBPS.