The amount of actual revenue collected by a hospital during - its rate year might vary from the amount approved in the preliminary cost base at the start of the rate year, since it is based on a series of assumptions and forecasts relating to volume, case mix, and other factors that could affect revenue. Therefore, at year's end, a final reconciliation takes place between the hospital and the Department of Health in which adjustments (in the next year's rates) are made to account for differences between actual net revenue and approved net revenue. OVERVIEW OF THE EVALUATION'S FINDINGS Almost immediately after the enabling legislation was in place, and the developmental work completed, a consensus began to emerge on the part of those familiar with the intricacies of the system that a comprehensive evaluation was needed. In response, the Health Research and Educational Trust of New Jersey assembled an evaluation team and raised the funds necessary to complete the work. In a general sense, the evaluation was aimed at increasing our understanding of the answers to four fundamental questions. First, is the system well designed and does it work as anticipated? Second, does the system make a difference in terms of the hospital's overall performance, effectiveness and efficiency in providing medical care? Third, what is the system's potential as a regulatory device, management information or data-based planning mechanism, and utilization review tool? And finally, on balance, what are the advantages and disadvantages associated with DRG reimbursement for hospitals, third-party payers, and others? From these basic questions a host of more specific issues and questions emerged. These were partitioned into nine separate analytical tasks which were undertaken by the evaluation team. [2] The detailed results of the team's analysis will appear in the following four volumes: Volume I: Introduction and Overview (published June 1982), Volume II: Economic and Financial Analysis, Volume III: An Examination of the Classification and Control of DRG Related Data, and Volume IV: The Politics and Organizational Impact of DRG in New Jersey. Although portions of the evaluation remain unfinished, a substantial body of work has been completed. Some of the more salient observations drawn from the analysis are presented below. It is important to bear in mind, however, that the findings reported here are preliminary in nature and therefore should not be subjected to the same level of scrutiny that might be appropriate for the final reports. A. Design Aspects Of The DRG System During the course of the evaluation, a considerable effort was devoted to assessing the actual way in which the system itself was designed. First, an analysis was conducted to see if there was any merit to the argument that the system didn't properly account for a patient's clinical status in assigning the patient to a DRG. Through the use of a Delphi survey of physi cians, hospital representatives, and researchers familiar with the DRG system, 37 of the 383 DRGs initially used to classify patients in 1980 were judged to contain patients whose assignments failed to adequately recognize differences in clinical status. Additionally, sixteen categories of problems which were deemed responsible for such assignments (e.g., placing patients with different illnesses in the same DRG) were identified. The effect of all this on reimbursement levels, however, is uncertain. A second aspect of the system's design, which was investigated in detail, concerned the structure and application of the reimbursement formula. The first type of analysis performed, regarding the formula's structure, was one whereby the various components of the formula were altered in an effort to assess the effect of these changes on DRG rates. Several observations can be made as a result of this process. First, the effect on DRG rates and direct care revenue of changing the "sharing factor" (set at .5 in 1980), which in part affects the proportion of the hospital's own cost and that of the statewide standard, was small. Next, when this factor was changed, the resulting effect on the incentive/ disincentive payments (i.e., the difference between the actual cost of treating the patient and the reimbursement rate) was proportionate to the change itself. And finally, as one would expect, the greater the reliance placed on the state average, the more the efficient hospitals were rewarded and the inefficient ones penalized. In an attempt to ascertain whether or not the formula was being applied correctly, a statistical analysis was conducted to see if there were variables that describe hospital characteristics, in addition to teaching status which has already been incorporated into the rate-making process, that should be included in the program, so as to improve the incentive structure. The evaluation team found that although some of the variables considered (e.g., size, location, etc.) did in fact lead to statistically significant reductions in unexplained variance in the dependent variables (i.e., cost and length of stay), the size of the reductions were small and therefore the additional complexity that would inevitably accompany their inclusion into the program could not be justified. Furthermore, the analysis confirmed that, in terms of both cost and length of stay, it is indeed appropriate to compute DRG rates separately for teaching and non-teaching hospitals. The next design factor studied was the way in which hospital costs were "mapped" into specific DRGs. The primary conclusion reached in this segment of the evaluation was that, with a few relatively minor exceptions, the cost accumulation, cost finding, and cost allocation processes used in the DRG system are consistent with traditional cost accounting definitions and concepts. It was Several specific problems, however, were identified. found, for example, that some of the terminology used in the 1979 Department of Health regulations is inconsistent with traditional cost and hospital accounting usage (e.g., employee fringe benefit . costs treated as a direct patient care costs). In addition, certain institutional costs considered to be unrelated to patient care (e.g., employee health insurance) should instead be regarded as being indirectly related. Lastly, the 1979 regulations de fined total direct care costs to be 100 percent variable and indirect costs to be 100 percent fixed. This doesn't accurately reflect the way in which these respective cost functions actually behave. It should be pointed out that many of these issues were addressed by the Department of Health in subsequent regulation development. B. Effect Of The System On Hospital Operations A second major area of investigation undertaken by the researchers centered on how the imposition of the DRG system affected the ways in which the hospitals function. First, using a matched sample of DRG and non-DRG hospitals, an examination of the impact of DRGS on the hospitals' organizational structures and processes was conducted. Briefly, the following observations were made: 1) decision-making authority in DRG hospitals is more decentralized than is the case in non-DRG hospitals; 2) the importance of the medical records department has increased dramatically in DRG hospitals due to the additional and more complex functions they've been called upon to perform, and the better skilled people required to execute them; 3) the medical staff in the DRG hospitals is more directly involved in hospital operations; 4) the quantity and type of information collected in DRG |