Organ Acquisition Cost Centers Part II: Reducing the Burden of Cost and Inventory

Transcription

1 American Journal of Transplantation 2006; 6: Blackwell Munksgaard Special Features C 2006 The Authors Journal compilation C 2006 The American Society of Transplantation and the American Society of Transplant Surgeons doi: /j x Organ Acquisition Cost Centers Part II: Reducing the Burden of Cost and Inventory M. Abecassis Feinberg School of Medicine, Northwestern University, Chicago, Illinois, USA Corresponding author: Michael Abecassis, mabecass@nmh.org Decreasing organ acquisition costs (OAC) reduces the cost of transplantation. The purpose of this article (Part II) is to build on the platform of definitions and concepts developed in Part I to achieve cost-savings in OAC. We propose several concrete strategies to decrease OAC. We will highlight the value of proactive waitlist management. We will develop the notion that waiting lists share significant similarities with product inventory and that inventory control and management can also decrease OAC. Finally, we will postulate that the value of the waiting list is determined by the number of recipients likely to be transplanted. We have coined the term transplantability index of a recipient waiting list. This index can be applied retrospectively to any waiting list in order to evaluate how actively the waiting list is being managed. This novel metric can also be used indirectly to assess and monitor waitlist inventory. Key words: Economics, kidney transplantation, resource utilization, waiting lists Abbreviations: OACC, organ acquisition cost center(s); OAC, organ acquisition cost(s); KACC, kidney acquisition cost center(s); KAC, kidney acquisition cost(s); OPO, organ procurement organization; SAC, standard (average) acquisition cost(s); ECD, expanded criteria donor(s); DCD, donation after cardiac death; LD, living donor(s); TI, transplantability index; LIFO, last in, first out; FIFO, first in, first out; FILO, first in, last out Received 27 June 2006, revised 29 August 2006 and accepted for publication 5 September 2006 Introduction Reimbursement of the proportion of OAC attributable to Medicare is made on a pass-through payment by Medicare to the hospital. Recovery of the proportion of OAC attributable to other payers is based typically on the balance between case-rate payments to hospitals that include OAC and the cost of the transplant procedure itself(1). Therefore, reducing OAC has a dual benefit. First, for Medicare beneficiaries, it decreases the cost of transplantation from a societal perspective. This benefits the tax payer. Second, and perhaps more important to the transplant centers, a reduction in OAC allows the transplant center to be more competitive with respect to the case-rate that they can accept from the commercial payers or payer networks. The standard (average) acquisition cost (SAC) for a specific type of organ (OAC) is determined by dividing the aggregate cost of evaluating potential recipients and living donors, listing fees, waitlist maintenance and organs procured for transplantation (numerator) by the actual number of transplants performed (denominator). Therefore, a shift upwards in the numerator or downwards in the denominator can result in a substantial increase in the SAC. In contrast, decreasing the numerator and/or increasing the denominator will result in decreases in the SAC. The numerator reflects the inventory; the denominator reflects revenues (Figure 1). The objectives of this article (Part II) are (a) to address specific strategies that will reduce essential cost components of OACC; (b) to develop the concept of the waiting list as inventory and to provide specific strategies that will reduce inventory and therefore waiting list costs; and (c) to introduce a novel metric that will assess and monitor the value of the institution s waiting list. Methods In August 2003, at Northwestern Memorial Hospital, we initiated a systematic analysis of OAC for the kidney transplant program. This included a bill-by-bill analysis of costs incurred in the evaluation, listing and de-listing process for potential recipients, as well as for potential living donors. We analyzed all vouchers submitted by various parties to the KACC. Monthly meetings were held with both professional and administrative personnel, representing both the hospital and the physician group, in order to address potential strategies to reduce KAC. Reduction of individual KAC components Our first effort was to make sure that all tests ordered and performed for both donor and recipient evaluations did not form part of routine healthcare for the individual. Medical societies have issued guidelines for screening for heart disease, colon cancer, prostate cancer, breast cancer, endometrial cancer, etc., for a well-defined population in each case. We implemented this measure immediately by not signing the prescriptions for these tests without asking whether the test should have been ordered by either the 2836

2 Organ Acquisition Cost Centers Part II Figure 1: Two hypothetical scenarios are described. In both scenarios, 200 patients are added to the waitlist annually. In scenario 1, 180 transplants are performed annually, whereas in scenario 2, only 80 transplants are performed annually. The incremental inventory accumulated over a 5-year period under both scenarios is illustrated. Note that potential recipients evaluated but not listed, and potential donors evaluated are not included in this figure, although they are clearly part of the numerator. primary care or referring physician as part of good medical care according to published guidelines. If so, we did not allow these tests to be billed to the KAC. This represented the single most important aspect of KAC reduction. For every type of evaluation that was allowable as a transplant evaluation (i.e. not for medical cause), a discussion with the provider of services was organized by the hospital administrator to ensure that the provider was discounting services to a reasonable level. At our institution, physicians from multiple disciplines who are involved in the evaluation of potential living donors and recipients have a common employer, the practice plan. Therefore, negotiations with the practice plan were held in order to decrease the percentage of charges billed by the physicians to the KAC. Similarly, negotiations were held with physician providers outside our institution and similar discounts were obtained. In an attempt to curtail administrative costs, a detailed analysis of hospital staff involved in pretransplant patient care was carried out utilizing benchmarking data, as well as assessing need. It was found that the programs were generally understaffed and adjustments were made. Staff salaries for local and regional transplant centers were obtained and analyzed in order to determine if salaries were comparable to those of other institutions and salary adjustments were made accordingly. In addition, the administrative staff responsible for reimbursing outside physicians and hospital service providers was instructed to reimburse only at a fraction of charges, similar to that negotiated with the physicians inside the institution. Also, a discussion was held with the immunology laboratory, which contracts with the hospital in an attempt to decrease the number and frequency of immunological testing as well as the cost per test. Finally, costs from outside laboratories for required blood work and other testing were targeted, as were costs from the organ procurement organization (OPO), transportation costs and all other costs involved in organ procurement from deceased donors. Reduction in inventory Recipients: We analyzed our kidney waiting list of over 800 recipients. We identified four populations of patients: (a) patients waiting who seemed appropriate; (b) patients waiting in whom we had not exhaustively approached the possibility for living donors; (c) patients waiting whom we had not assessed for several years; and (d) patients waiting who were no longer considered candidates but who were still on the list, some of whom had been inactivated. We also analyzed our intake process for new evaluations. We categorized them by using color coding: (a) green: excellent candidates, no apparently obvious contra-indications; (b) yellow: reasonable candidates with multiple medical problems, presumably not constituting absolute medical contraindications; and (c) red: poor candidates, likely to have absolute contraindications to transplantation. We also looked at other specific parameters that we felt would potentially increase the likelihood of transplantation, such as the availability of recent blood received by the organ bank for cross-matches, and the percent of patients on our list that had consented to accepting kidneys from expanded criteria donors (ECD). Living donors: We analyzed our intake process for potential living donors. We started to match potential living donors to the likelihood of a successful evaluation in the recipient. We streamlined our evaluation process, but structured the process into four distinct phases: (a) initial information session; (b) submission of health care questionnaire and blood type; (c) formal medical and psychosocial evaluation, including blood work and CT scan if the history and physical examination by the independent advocate were judged to be appropriate for living donation; and (d) information session leading to informed choice and informed consent. There were two objectives here: first, to avoid unnecessary testing in donor candidates who failed phase 2 and the first portion of phase 3 and, second, to avoid considering multiple living donors simultaneously or potential living donors for recipients who we did not feel would be considered appropriate recipient candidates. Results Although the strategies described above are a work-inprogress at our institution, by implementing some of the measures discussed above, we were able to reduce our KAC by 19% in 2003, 5% in 2004 and a further 14% in 2005 for a total reduction of 34% over this 3-year period. Some of the measures had an immediate impact on KAC while others resulted in longer-term reductions. The following data are meant to illustrate the concepts put forth in this report. Individual KAC Between 2003 and 2004, we were able to reduce recipient evaluation costs (net of professional fees) by 78% and were able to maintain this reduction in These costs included blood work and procedures performed on American Journal of Transplantation 2006; 6:

3 Abecassis Table 1: Reduction in individual components of KAC. Reductions in KAC were achieved by implementing measures with the various service providers Service provider Services provided Reduction % Reduction All Not permitted (i.e. for medical cause) Yes 78 Physicians within institution Donor and recipient evaluations Yes 40 Providers outside institution Donor and recipient testing Yes 40 Outside laboratories Blood work, serologies, etc. Yes >50 Immunology laboratory Tissue typing, cross-matches Yes 5 Staff within institution Coordination of care: pretransplant No N/A OPO Organ procurement costs No N/A N/A = not applicable. potential recipients that could otherwise be performed for medical cause and billed to the recipient s insurance. In the past, these evaluations had been billed to KACC as part of the transplant evaluation. This was justified insofar as these evaluations were necessary for listing decisions. This was clearly the most dramatic reduction in KAC related to the strategies implemented. In terms of the allowable costs (i.e. strictly needed because a transplant was contemplated and not billable for medical cause), we reduced physician costs within the institution through a straight 40% discount from charges constituent with the practice plan s average collection rate of 60%. A similar reduction was also achieved for services provided by physician groups outside the institution. We made a decision that all blood work performed at outside laboratories would be reimbursed at Medicare allowable rates. This resulted in substantial reductions in these costs. We were unable to achieve a reduction in costs related to the institutional staff responsible for pretransplant services mainly because, according to benchmarks and to our own observations, we were understaffed in certain categories, such as social work and nutritional support, and therefore, we made appropriate adjustments in these. We also carried out benchmarking efforts, comparing staff salaries to those matching their job descriptions at both local and regional institutions, and found these to be at market value for the most part. We made necessary adjustments to ensure parity where necessary. In negotiating with the immunology lab, bids were requested from competing laboratories and appropriate negotiations were carried out. This process is ongoing with further adjustments pending. Also, we were able to revisit the type and frequency of testing and adjust these, while continuing to be compliant with the relevant regulations. We were unable to achieve cost reductions with the OPO. Table 1 illustrates the results from these efforts. Inventory We instituted measures to proactively manage the waiting list in order to reduce inventory costs. We decreased the numerator by managing the listing and de-listing processes. We expressed to these patients a more accurate assessment of their health. Minimal evaluations were performed on potential recipients who were felt to have obvious contraindications to transplantation (red or yellow see Methods). Also, we removed (de-listed) candidates from the waiting list who had absolute contraindications. We found that 27% of our waiting list had no recent blood samples forwarded to the organ bank. However, we found that a lot of these patients were either inactive or were no longer considered candidates, yet they remained on the list, some with an active status, others with inactive status. We implemented a program to address this by contacting those eligible for transplantation without fresh blood samples in the organ bank and by de-listing those who were no longer considered candidates. Altogether, we achieved a 15% reduction in our inventory in 2004 and a further 1% decrease in Another strategy to reduce inventory is to increase the relative number of living donor transplants or, alternatively, by expanding the deceased donor pool (denominator). The first goal can be achieved by encouraging living donation at the initial recipient evaluation and by revisiting the possibility with recipients on the waiting list. Living donor transplants are associated with better overall outcomes. However, this strategy can increase the costs of LD work-ups unless processes are implemented to (a) limit work-ups of cases when the recipient is unlikely to pass their own evaluation, (b) limit the number of LD evaluated simultaneously, and (c) make sure that LD who are deemed ineligible in the first two phases of evaluation do not proceed to the third phase. We achieved a 6.7% reduction in LD costs with the use of these strategies despite a 7% to 12% annual increase in LD between 2002 and The second goal can be accomplished by the increased use of marginal donors such as ECD and donation after cardiac death (DCD) donors. Also, recipients need to be educated regarding ECD and DCD donors and, where appropriate, they should be consented. We found that only 10% of our waiting list had consented for ECD and we are in the process of educating our potential recipients about this possibility. Table 2 illustrates the results of these efforts. Transplantability index (TI) Parallel with our efforts to reduce inventory costs, we developed a metric that we have coined the transplantability 2838 American Journal of Transplantation 2006; 6:

4 Organ Acquisition Cost Centers Part II Table 2: Reduction in inventory. Various strategies were implemented leading to reductions in the number of unnecessary work-ups and in the waitlist inventory. The table illustrates reductions achieved in many areas. Overall, we were able to achieve a 16% reduction in inventory. Also, despite an increase of 7% and 12% in LD over the study period, we were able to achieve a 6.7% reduction in LD costs Element Action Reduction % Reduction 1. Recipients Potential LD Re-evaluate (LD) Yes 10 Questionable candidate Re-evaluate Yes 10 Poor candidate De-list Yes 5 Blood for cross-match Educate Yes 5 ECD consent Educate No N/A 2. Living donors Recipient candidacy Determine Yes 10 Number tested Limit Yes 10 Phases Avoid later phases Yes 5 Table 3: Calculating the TI: the table illustrates the method used to calculate TI for our institution. The estimated number of transplants based on the list demographics, average waiting times, the ratio of the institutional waiting list to the OPO waiting list and the available kidneys from deceased donors in the OPO. The actual number of transplants performed is then divided by the estimated number of transplants in order to determine the transplantability index of the list. The TI can be used to monitor practice patterns that may impact the TI year-to-year at the same institution TI = Number of of kidney transplants performed with deceased organs in center [ (number of recipients on list per blood type) (1/average waiting time for blood type in years) (% recipients with PRA < 50)/[(total number of recipients in OPO waiting list) (1/average wait time for all blood types in OPO)] (number of deceased donor kidneys transplantedin OPO)] Table 4: Calculation of TI at Northwestern: following the formula outlined in Table 3, we utilized our institution s known values for CY to calculate TI No. of kidney Center OPO Year transplants waitlist waitlist TI CY CY CY index of the waiting list. The TI is a ratio of the actual number of kidney transplants from deceased donors performed at a particular institution within a defined time period, to the number of transplants, also from deceased donors, that should have been performed given a few demographic parameters of the institutional waiting list and the availability of deceased transplantable organs in the OPO. The demographics of the waiting list can include the level of sensitization [recent panel reactive antibody (PRA)], blood type and average waiting time. The number of available deceased donors in the OPO and the ratio of the institution s waiting list compared to that of the other transplant centers in the same OPO determines the expected institutional share of the deceased donor pool. Based on efforts described to reduce inventory, the TI at our institution increased from 0.87 to 1.03 between 2003 and Tables 3 and 4 outline the calculation of TI. Discussion Transplantation saves thousands of lives in the United States every year. Every attempt should be made to increase the number of transplants and, therefore, the number of lives saved and to improve the quality of life in patients receiving transplants. Clinical outcomes cannot be compromised by any financial consideration. Nonetheless, the rising costs of healthcare are evident in every discipline of medicine. Transplant professionals are becoming increasingly aware of financial issues surrounding transplantation and these cannot be ignored. In this report (Part II), we have introduced some theoretical constructs that provide a financial framework for reducing OAC and have presented specific strategies for decreasing OAC. We have provided a clear case study of the value of proactive OAC management and containment. We have proposed and evaluated concrete strategies to reduce OAC based on limiting evaluations to allowable services at a reasonable cost as defined in Part I. We have also proposed several strategies to reduce the waiting list inventory. There are great similarities between product inventory and transplant recipient waiting lists. In the manufacturing industry, inventory costs are viewed as a significant threat to net operating income (profit). If a widget factory escalates production of widgets, at first this is viewed as a positive indicator of productivity and profitability. But what if no customer orders exist for the widgets? In a balance sheet, a type of financial statement, the widget inventory will appear as assets and will be assigned market values. But, in another type of financial analysis, a statement of cash flow, inventory is not associated with American Journal of Transplantation 2006; 6:

5 Abecassis revenue, negatively influencing the bottom line (net operating income). It is generally accepted that large inventories result in negative margins (2). Therefore, just-in-time production, as demonstrated by Toyota several years ago, limits the accumulation of inventory (3). Living donors and those with a green designation (see Methods) represent the best examples of just-in-time transplants and are the best illustration of inventory churn. We have demonstrated that the implementation of a few simple measures can result in a reduction in waiting list inventory. We did not show a decrease in inventory costs because we did not measure these directly. Nonetheless, it is clear that proactive waitlist (inventory) management can reduce OAC. Accounting principles normally used in measuring and managing inventory can be applied and would be particularly applicable to transplantation. For instance, last in, first out (LIFO) and first in, first out (FIFO) would apply specifically to recipients of living and deceased donors, respectively. Just-in-time transplants would thus constitute LIFO inventory (4,5). We have defined the transplantability index (TI) of a waiting list as a novel metric that can be used to assess and monitor the elements of practice patterns at an individual institution. The application of this metric is limited to centers at OPOs that service more than one transplant center. This metric can monitor efforts aimed at reducing waiting list inventory. TI should equal 1.0 if the transplant center s waitlist is undergoing transplantation at a rate equal to that of the other centers in the OPO. If waitlist management is suboptimal or if the other centers in the OPO are managing their lists more efficiently, the TI will fall below 1.0. In contrast, if the center is managing the waitlist efficiently so that inventory is decreased relative to the other centers, the TI will exceed 1.0. In theory, as inventory is decreased, the TI should increase. Waiting time is currently the most important determinant of priority status, although PRA is an important determinant of transplantability irrespective to waiting time. Any change in allocation policy is likely to affect the TI directly. However, the formula for TI would change as new allocation policies are developed, including a consideration of net lifetime benefit, but the concept would remain the same (6). The TI might change yearto-year at a particular center only if the demographics of the waiting list or the number of available organs from deceased donors change, or if the transplant center suddenly changes its practice patterns. The TI is likely to provide an excellent tool for individual institutions to assess and periodically monitor waitlist inventory. A similar concept could be applied to nonrenal organs. The components of the TI would be different, depending on the allocation principles. Thus, for liver transplantation, the MELD score would be a major determinant of the TI (7). Conclusions Decreasing OAC is a worthwhile effort for a variety of reasons alluded to in this report. Active waitlist management results in reduced costs and improved clinical outcomes. Several strategies for achieving this goal have been proposed that include (a) monitoring and control of the various OAC components, and (b) a reduction in waiting list inventory. A novel metric, the TI, has been defined for patients awaiting kidney transplantation. This measure can also help assess and monitor practice patterns at individual institutions and waiting list inventory. Further studies are needed to validate these findings across organs and in other institutions. Acknowledgments I would like to acknowledge Gwen McNatt and Luke Preczewski for their valuable contributions and helpful discussions of the manuscript and Dr. Dixon Kaufman for his review of the manuscript and suggestions. References 1. Abecassis MM. Financial outcomes in transplantation A provider s perspective. [Minireview] Am J Transplant 2006; 6: Muller M. Essentials of inventory management. New York: American Management Association, 2002: Monden Y. Toyota production system: An integrated approach to just-in-time, 3rd Ed. Woodstock, CT: Spring Publishing; Morse D, Richardson G. The LIFO/FIFO decision. J Account Res 1983; 21: Dopuch N, Pincus M. Evidence on the choice of inventory accounting methods: LIFO versus FIFO. J Account Res 1988; 26: Winkelmayer WC, Weinstein MC, Mittleman MA, Glynn RJ, Pliskin JS. Health economic evaluations: The special case of end-stage renal disease treatment. Med Decis Making 2002; 22: Axelrod DA, Koffron AJ, Baker T et al. The economic impact of MELD on liver transplant centers. Am J Transplant 2005; 5: American Journal of Transplantation 2006; 6: