Diagnosis related group-based cost minimization models: applications in a hospital environment

The primary objective of this article is to investigate 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 94,500 observations (1999 and 2000), obtained from a teaching hospital in South Florida. The top ten Diagnosis Related Groups (DRGs) with the highest volume are selected and classified into four insurance categories: Medicaid, Medicare, commercial, and self-pay. The cost models are fitted to the data for an average R² value of 79%, and a MAPE value of 15%. 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 6,367 admissions for the selected DRGs in 2000, the total cost per year and the cost per patient day decreased by approximately 11.58% and 10.35%, 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 costs. In addition, this research offers a decision support instrument for healthcare administrators.