Considerable attention has been devoted in the accounting literature to identify the factors that cause or drive the costs of overhead activities. This paper extends recent cost driver research to the health care provider. In various case studies, it has been suggested that overhead costs are driven by volume and complexity variables. This paper investigates the significance of these variables in determining hospital overhead costs, how they are structurally related and how the cost impacts of these variables can be estimated in practice. This paper analyzes the determinants of hospital costs using the sample of South Korea hospitals for seven year during the period 1952－1997. The paper focuses on the extent to which hospital overhead costs depend on complexity, efficiency in addition to depending on more conventional volume based measures of hospital activity. The results of regression analysis suggest that volume and complexity factors positively and significantly affect overhead costs in the hospital industry. The results show that the complexity-related cost drivers strongly affected on the overhead costs in tile health care provider industry more than manufacturing industry which is mainly affected by volume-related cost drivers. That means each Industry may have different cost structures. Therefore it Is Important to find their proper cost structures and cost drivers and use them. Futhermore identification of overhead or indirect cost drivers is likely to be particularly useful in heath care. The identification of cost drivers can be of benefit to all health care stakeholders because these facilitates more efficient management of the national resources devoted to health care. While this study has documented that the level of service complexity is a significant determinant of hospital overhead costs, caution should be exercised in interpreting this as supportive of the cost accounting procedures associated with ABC. It is an open question whether even a well-designed ABC system will provide suitable proxies for marginal costs for decision making purposes.