Twelve regional hospitals voluntarily participated in this study. There regional hospitals were equipped individually with 500 beds or more and were located in six regions. Date colleted were from July 2004 to December 2004, for analysis of mean medical expense, mean hospital days, expenses of Diagnosis Related Groups (DRGs) cases paid and distributed section of ABC for DRGs different weight project, CMI, base rate coefficient of variation (CV) of the expenses of DRGs items for DRGs payment project. Then we may re-evaluate the influence of DRGs system promoted by NHI on regional hospitals. Analyses revealed that according to the payment percentage schedule, the three payment modes with different exclusions, the estimated reduction to regional hospitals is 15% as the highest. Reimbursement to these twelve hospitals vere reduced by 164 million points to 415 million points in one quarter, and the mean reduction for each hospital was 14 million points to 35 million points quarterly, on the average. For the top ten DRGs items, the negative sum difference is between 5 million points to 17 million points. The probable causes are that more than 50% of the DRGs items are in B and C section, the scheduled point value is too low; and the critical points for expenses and deviated. This study also revealed that variation within the DRGs group item based on the hospital’s characteristics, disease severity, and teaching costs, in order to seek reasonable rates for individual items and to confront the impact caused by the DRGs system in time while the current planned payment of DRGs are unable to explain comprehensively the variation of expenses in practice.