مدل به حداقل رساندن هزینه برنامه های کاربردی در یک بیمارستان آموزشی

The primary objective of this article is to explore the feasibility of the application of cost minimization analysis in a teaching hospital environment. The investigation is concerned with the development of cost per admission and cost per patient day models. These models are further used for determining the value of the length of stay that would minimize cost per patient day (projected length of stay) and for estimating the costs. This study is based on total of 109,060 observations (2002), obtained from a teaching hospital in South Florida. The top 10 diagnosis-related groups (DRGs) with the highest volume are selected for the study. The cost models are fitted to the data for an average R2 value of 87.3%, and a mean absolute percentage error (MAPE) value of 16.1%. The result demonstrates that if a hospital can control the length of stay at the projected level, on average, the cost per admission and the cost per patient day will decrease. Based on 8703 admissions for the selected DRGs in 2002, the total cost per year and the cost per patient day are decreased by approximately 8.56% ($15,453,841) and 4.02% ($66.30), respectively. Overall, these results confirm that the concept of cost minimization analysis in economic theory can be applied to healthcare industries for the purpose of reducing of costs. Cost minimization and cost variation analyses offer useful information to hospital management for better decision-making. It would be an important aid in making management decisions, particularly for cost reduction.

The last three decades various economic models have seen applied for the healthcare industry. There have been numerous studies on the topic, nearly all in hospital environment (Tousignant et al., 2003, Smet, 2002, Dawson, 2002, Cleverley, 2002, Kearns et al., 2001 and Jackson, 2000). Most of the models are generally focused on cost structure, cost-effectiveness analysis, cost utility analysis, and cost-benefit analysis. Although profit maximization has traditionally been assumed in the empirical analysis of many other industries, a variety of both institutional and behavioral features of the hospital suggest that the assumption of profit maximization is not appropriate in the case of hospitals (Cowing et al., 1986). Cost minimization is a much more reasonable assumption than profit maximization for various non-profit organizations. Robinson (1993) notes that cost minimization analysis is an appropriate technique to use for healthcare organizations when there is reason to believe that the outcomes of the different patients’ length of stay under consideration are the same. There have been countless applications of cost minimization analysis for healthcare organizations (Jones et al., 1999, Shepperd et al., 1998, Coast et al., 1998 and Halvorsen and Kristiansen, 1996). According to these studies, the cost minimization analysis was performed to determine the least costly option among several alternatives (treatment strategies) that will yield an equivalent outcome. Nevertheless, very few information is available regarding the application of cost minimization to determine the optimal length of stay that lead to minimize cost for the diagnosis-related groups (DRGs) system. A significant study which applies the cost minimization analysis for the DRGs at a teaching hospital, was conducted by Suthummanon and Omachonu (2004). According to this study, the DRGs were further classified into four major insurance categories: Medicaid, Medicare, commercial, and self-pay. Regarding to this study the results confirm that the concept of cost minimization analysis in economic theory can be practically applied to healthcare industries for the purpose of reducing of costs. A shortcoming of this work is that the analysis for other insurance types and overall patients as a whole (all insurance types are grouped together as one model) is ignorance. Due to there are several insurance types in each DRG, therefore it is very difficult to individually analyze for each one. Hence, analysis for overall patients should be performed with the aim of understanding the cost behavior as a whole. The literature on hospital cost models encompasses a wide variety of models. At one extreme, there are models that treat costs as dependent variable and employ numerous input factors as independent variables. The examples of widely used independent variables are patient’s age, gender, race, marital status, income, health condition, and health insurance. At the other end, there are structural cost models in which costs (or the logarithm of costs) served as the dependent variable, and outputs are used as independent variables. Berki (1972) states that the cost model can be presented as TC = f(Q), where TC denotes total costs and Q the quantity of output. It is important to note that patients’ length of stay is used as the quantity of output in many healthcare studies ( Custer and Willke, 1991 and Cassimatis, 1988). Several studies have focused on cost function using the length of stay as an independent variable. In addition, previous studies including Carr and Feldstein, 1967, Francisco, 1970, Dowling, 1976, Elnicki, 1979, Grannemann et al., 1990, Phelps, 1997 and Malkin, 1998 conclude that the length of stay is the most determinant of hospital costs. The primary objective of this research is to examine the feasibility of the application of cost minimization analysis for DRGs at a teaching hospital. The investigation attempts to determine the level of the length of stay at which it becomes more cost-effective to provide the services and to examine potential cost saving if the projected length of stay has replaced the actual level. The two questions addressed here are: (1) what is the projected length of stay to minimize cost per patient day? and (2) what are the effects of changing the length of stay from the actual level to the projected level, on the total cost and the cost per patient day for a hospital?

This analysis is conducted at a large teaching hospital in the United States. The results verify that the concept of cost minimization analysis used in economic theory can be practically applied to a healthcare environment. However, there are a large number of factors that affect the cost behaviors at different hospitals. These factors can be both complex and wildly different from hospital to hospital. Furthermore, different hospitals have different characteristics (teaching status, size, ownership, location, etc.), which may affect cost behaviors. Thus, the cost per admission models, which show the relationship between cost and length of stay, might be constructed as linear (or a log transformation), quadratic, or cubic functions. Still, the similar analysis may be applied to these healthcare organizations. By comparing these results with the work conducted by Suthummanon and Omachonu (2004), the outcomes are alike. In other words, the projected length of stay would lead to minimize the cost per patient day and to reduce the total cost per year. This implies that the cost minimization analysis theory can be applied for the DRG systems in a hospital. Moreover, the results suggest that there are comparable outcomes between the analysis under individual insurance type (previous research) and using overall patients (all insurance types are grouped together as one pool). However, the results of the cost function and the amount of cost saving might have some variations. Cost minimization and cost variation analyses offer useful information to hospital management for better decision-making. The projected length of stay obtained from the cost minimization analysis provides a level of length of stay that would minimize the cost per patient day. The results of the cost variation analysis demonstrate that if a hospital can control the length of stay at the projected level, on average, the cost per admission and the cost per patient day will decrease. Particularly, if a hospital can control the length of stay at the projected level, the cost per patient day will be minimized. It would be very helpful for policymakers to have more information about the costs and the length of stay as they consider efficient management. The health administrator could use this information to work with medical staffs to determine any alternative treatments or adjust the treatment patterns in an attempt to control the length of stay (close to the projected length of stay). Thus, the ability to alter the length of stay (close to the projected length of stay) leads to a decline in healthcare costs. A financial incentive approach such as profit sharing may be use to encourage the medical staffs in this effo