Cost Containment

In the accompanying article by Dexter and Tinker [1], the authors have conducted a desktop study, postulating what financial savings might be accrued if serious anesthesia-related mishaps were avoided in four types of surgery--kidney transplant, total hip replacement, laparoscopic cholecystectomy, and cesarean section--at a single university tertiary health center. In addition, the authors have postulated what savings might be achieved if the expenses for the most costly 10% of cases for those four types of surgery plus two additional surgeries, lung lobectomy and coronary artery bypass, could be reduced to no more than the median level of expenditure for each of those surgeries. Following the latter calculation, an estimate was made of the potential percentage increase in the median expenditure possible for all patients if the funds from the savings achieved by lowering the excess expense above the median for the highest 10% of patients were distributed among the other patients. For the first four types of surgery, the authors conclude that the difference between the costs for the patients with anesthesia-related complications detected by the retrospective epidemiology nurse chart review process and the median cost for care for other patients in the same surgical categories was such a low percentage of the total costs (<or=to2%) for all patients undergoing those surgeries during the time period of the study that the savings were insignificant. They imply therefore that there would be no financial benefit to any further improvement in anesthesia "quality" for these surgeries. Second, for those first four types of surgery, elimination of the highest 10% of cases only reduced total costs by 8%-10%, which would have allowed an increase of 8%-11% in total costs for all patients, not very much to spend for significant improvements in the authors' opinion. On the other hand, for lung lobectomy or coronary artery bypass procedures, elimination of the highest 10% of cases reduced total costs by 30% and 16%, respectively, which would allow increases in total expenses by 45% and 19% for the other patients, presumably a worthwhile dollar amount to utilize for improvements. These types of conclusions raise interesting issues in an economic climate that is becoming ever more limited by cost considerations. In a health care institution where the routine level of care is of high quality and for procedures where complications attributable to anesthetic management of an immediate and serious nature are infrequent, the argument is made that additional expensive efforts applied to all patients in attempts to eliminate rare, nonnegligent complications for the few cannot be justified from a cost perspective. Although the theory is not tested in this study, the authors propose that overall cost savings for surgeries that are already safely medically managed with current techniques and monitoring methods must be achieved by improved efficiency in management of time, capital and expendable resources, and personnel within the specific health care environment (e.g., case management, critical path methodology, care map approach, etc.). For lung lobectomy and coronary artery bypass surgeries, the authors did not analyze the incidence of anesthetic complications, although by inference the reader must assume that the most expensive 10% of cases were a result of complicated medical outcomes in general. In the latter two groups of patients, the substantial variation in resource consumption between the most costly 10% of patients and the median resource consumption lead the authors to conclude that improvements in the quality of medical care per se that presumably would impact intensity of care required and/or length of stay for these patients would be fiscally justified. However, the authors do make clear that such a fiscal analysis suggests that alternative therapy to surgery would be the most efficient way to control costs for those patients in whom complicated perioperative courses are likely despite the highest quality care presently available. It must be stated that our society is presently struggling with the latter concept, as high-risk patients have been able to win legal judgments supporting their demands for extremely costly medical treatments. What are the limitations in these data and limitations in these interpretations that should be considered? First, relative risk is not defined in this work and the term "low risk" appears misapplied. The Methods section states that lung lobectomy and coronary artery bypass are "high-risk operations," while the other four procedures were chosen "because they are from different surgical specialties." Neither kidney transplant, total hip replacement, laparoscopic cholecystectomy, nor cesarean section is "low risk" compared to many peripheral procedures, which is corroborated by the complications the authors recorded in this group, including several myocardial infarctions, transfusion reaction, sepsis, pneumonia, aspiration, hemomediastinum, and two deaths. It would appear that truly "low-risk" procedures were not included, nor were anesthetic outcomes of relevance to low-risk procedures examined in this study. For truly low-risk procedures, minimization of anesthetic sequelae such as inadequately treated pain, excessive sedation, or untoward nausea and emesis may decrease hospital costs if patient throughput is enhanced, fewer staff are needed, and fewer postoperative resources are consumed. Furthermore, the "cost efficacy of hypothetically eliminating adverse anesthetic outcomes" was never applied to the two "high-risk" procedures, lung lobectomy and coronary artery bypass. The authors state that they were unable to separate the surgical from the anesthetic causes of morbidity in the lung lobectomy and coronary artery bypass patients. Hypothetically eliminating the most costly 10% of patients undergoing lung lobectomy or coronary artery bypass is not necessarily equivalent to eliminating the costs of consequences of adverse anesthetic outcomes, since the extra expenses may have occurred despite excellent anesthetic care. Second, the costs analyzed in this study are not true costs per se, but rather "costs" backed out from hospital charges by using a fixed "cost to charge" ratio applied to the gross hospital charges adjusted by a multiplier representing the overall percentage increase in gross charges between the occurrence year and 1993. Few hospitals actually tabulate true costs of actual wages, supplies, overhead, etc. Percentage mark up for different items and services varies, "OR time" charges cover numerous bundled expenses, charges for laboratory tests may subsidize other less profitable departments, charges for indirect expenses can include items such as institutional indebtedness, etc. Thus calculating "costs" as a percentage of the overall hospital charges may be far removed from determining actual resource consumption if the environment were free of reimbursement concerns. Third, the data on complications were not prospectively collected with the purpose of this study in mind and were not collected by individuals specifically concerned with achieving the aims of this study. Retrospective collection is limited by the data preserved in the medical record, which often is incomplete, as well as by the tenacity of the investigators to pursue any and all clues available in such records. Fourth, this work takes only the view of the surgical facility with regard to cost-benefit analysis. The assumption is made that societal costs are equivalent to hospital "costs" as determined by the imperfect "cost to charge ratio" methodology. The study was not able and was indeed not designed to assess fiscal implications of post-discharge medical resource consumption, lost wages, lost productivity for the employer, additional costs of alternative care for dependents, requirement for house-hold help, etc. For shorter-stay procedures, improvements in anesthetic quality that return patients promptly to their baseline physical and mental status may yield significant cost savings to the individual, the family, the workplace, and the community that may not appear to be worth pursuing if one accepts the inference in this article that anesthetic quality has plateaued for less than high-risk procedures as far as cost-benefit to the hospital is concerned. The methodology in this study also underestimates some societal costs such as medicolegal costs and disability costs resulting from outcomes such as death or myocardial infarction, since death may actually reduce hospital "charges" in the short run. If medicolegal and disability costs are included in the analysis, seemingly expensive adjuncts to improve anesthetic quality/reduce adverse outcomes even for the moderate-risk procedures may be cost effective overall. Fifth, only when care is excellent is there presumably no benefit to improving it. No guidelines are provided for readers to determine whether care given in any other particular setting is "excellent." Was the care rendered in generating the data for this analysis indeed optimized? The only information provided is that the incidence of two complications, spinal headaches and myocardial infarctions, in these relatively small numbers of patients was similar to values previously published in textbooks. Finally, it is possible that the 2% reduction in "costs" achieved by eliminating serious anesthetic adverse outcomes for the four moderate-risk procedures, which the authors conclude was a minor savings, may not appear paltry in dollar terms versus percentage values if the numbers of patients to which the percentage savings pertains is great. In summary, readers need to appreciate that many aspects of this study are conjectural, such as the premise that expenses for the costly patients can be "divided up" and reassigned as extra allowable expenses for other patients. Other assumptions detailed above have been made that limit the interpretation of the results and applicability of the conclusions. The significance of this hypothetical work therefore is principally its thought-provoking suggestion that major investments of costly resources to further reduce procedural risk should be principally directed at selected procedures with known extreme variability in outcomes, including potential consideration of alternate therapies for those patients with a higher likelihood of adverse outcomes who may be contemplating high-risk surgeries. From the health care institution fiscal perspective taken in this article, if the level of care is excellent, the focus for the majority of surgical patients should be attention to system-wide improvements in efficiency of resource utilization, including personnel, space, time, consumables, and capital investments, in order to produce further additional overall cost reduction per good patient outcome.