*Abbreviations: AAC, average accumulated costs, CAD, cadaveric, DRG, diagnosis-related group, ESRD, end-stage renal disease, HCFA, Health Care Financing Administration, LD, living donation, USRDS,United States Renal Data System.
The federal government has been the main payer for renal transplantation for several decades. Medicare payments for renal transplantation covered end-stage renal disease (ESRD *) as part of the Social Security Amendment of 1972, while Medicaid coverage came later (1). Private third-party payments and heavy subsidies in research account for the remainder. Immediately after Medicare’s initial coverage, hospital transplant charges and costs were not easily accessible or clearly defined. With managed care and changing health care policy in the 1980s and 1990s, focus has centered on identifying and understanding payments, costs, and charges in many aspects of health care delivery, including renal transplantation (2–5).
Renal transplantation is currently accepted as the most cost-effective therapeutic option to treat ESRD (6). Historically, however, the comparison has been between cadaveric transplantation and hemodialysis (7). At some point, usually several years after transplantation, the net cost savings of renal transplantation are realized. Much of the difference in costs has been attributed to increased morbidity on dialysis.
By the late 1980s, living related renal transplantation was a well accepted option for patients with ESRD. By the mid 1990s, higher graft and patient survival rates were noted comparing living graft recipients to cadaveric graft recipients (8). Additionally, grafts from HLA-identical siblings had the highest survival, followed by spousal and living unrelated donors, followed by cadaveric grafts (8). The reasons for the higher survival rates are many. It appears that living donors will comprise a significant portion of the donor pool in the years to come.
With a framework for renal transplant cost analysis already provided, and a living donor pool with longer graft survival, it seems intuitively obvious that there are cost savings associated with living donor renal transplantation compared to cadaveric donation. Possible sources for cost savings have included lower costs for avoiding delayed early graft function and primary graft failure, less immunosuppression, and even the financial implications of improving the productive capacity (“opportunity costs”) of recipients who are off dialysis (9,10). Interestingly, given Medicare’s current policy of reimbursing equally for cadaveric or living related renal transplantation, the cost savings could have policy implications at the federal, state, and local level. This study identifies some of the sources of the additional cost savings from living compared to cadaveric donor renal transplantation and identifies certain beneficiaries of this financial advantage.
The data used in this study came from the United States Renal Data System (USRDS) (11). The USRDS is a joint effort of the National Institute of Diabetes and Digestive and Kidney Diseases and the Health Care Financing Administration (HCFA). It was designed to collect, analyze, and distribute data describing ESRD in the United States including prevalence, treatment modality, survival, and cost of care. The USRDS provides linked records for all transplants performed in the United States from the United Network for Organ Sharing renal transplant registry and the HCFA’s billing and payment records. This enables linkage between the characteristics of transplant patients and detailed information on the charges and payments for the medical services provided to them.
Between 1991 and 1996, information was available for 42,868 CAD and 13,754 LD transplants. More than 5 million Medicare payment records were analyzed. We calculated the difference in average payments made by Medicare for CAD and LD for services provided during the first posttransplant year in the following eight categories: inpatient hospital, outpatient hospital, dialysis, clinic/home health, surgeons, laboratory & radiology, immunosuppression, and other medical.
Payments and charges.
In a second analysis, we looked at profitability in terms of payment to charge ratios for CAD and LD. In order to select only those patients who had Medicare as a primary payer, we used a cut-off of at least $15,000 in Medicare payments for the technical component of the transplant hospitalization as an inclusion criteria. Some patients with Medicare as the primary payer may have been excluded; however, the proportion of Medicare primary patients included in this sample should be high.
Average accumulated “costs” (AAC) were calculated with a generalization of the Kaplan-Meier methodology to continuous data such as cost (12). This information preserving methodology uses data on patients with incomplete follow-up by systematically accounting for the censoring of observations at the time of last known follow-up. Specifically, the AAC for the first day after transplant equals $0 as no patient is discharged that quickly. For each subsequent day “t,” the average accumulated cost equals the average accumulated cost of the previous day plus the average costs incurred on day t (AICt). AICt was calculated as the total costs incurred on day t (TICt) by the individuals remaining uncensored on day t (patients with follow-up at least up to day t) divided by the number of patients uncensored on day t (nt):AACt= AACt−1+ (TICt/nt), for all t >1.
Theoretical standard errors are not available for the information measured by the accumulated cost methodology used here. However, bootstrap methods are available to produce very accurate estimates of error and confidence intervals by randomly resampling from the original sample (13). Statistics of interest generated by these resamplings provide information about the distribution of the parameter being estimated.
A matrix of weights was filled randomly for 2000 replications. Bootstrapping was done by repeating the algorithm for calculating the accumulated costs and their differences with a different weighting system each time. The distribution of the bootstrapped statistics gives confidence intervals as well as significance levels for individual results (14). All P-values were from two-tailed tests and were calculated at discrete points in time (i.e., total cost to 1 year after transplant, and total cost of the transplant hospitalization).
Average total Medicare payments for all kidney transplants in the US were $39,534 and $24,652 for CAD and LD, respectively.
This $14,882 difference in Medicare payments for the first posttransplant year was statistically significant (P <0.0001). These average total payments made by Medicare did not include organ acquisition costs.
Dominant source of the difference in average Medicare payments for all kidney transplants in the US came from hospitals.
The $10,653.67 difference (P <0.0001) in inpatient hospital payments accounted for most of the total difference (Fig. 1). These inpatient payments included the transplant stay, immunosuppression, and other transplant services. Rates of rehospitalization and the most common diagnosis-related groups (DRGs) indicated for hospitalization are presented in Table 1. The average number of rehospitalizations was 2.51 for CAD and 2.28 for LD (P <0.0001). The five most common DRGs had similar prevalence and because of the large numbers were significantly different (P <0.0001). Significant differences were also observed for outpatient hospital payments ($692.61), dialysis ($814.98), clinic/home health ($406.13), surgeons ($587.35), laboratory and radiology ($405.76), immunosuppression ($442.68), and other medical ($985.31). All of these differences were significant (P <0.0001) (Fig. 1).
Total charges and payments for the first 5 years after transplant for all kidney transplants in the US were higher for cadaveric versus living donation.
All charges and payments made during the first 5 posttransplant years (1991–1996) were recorded. Five-year cadaveric donor transplant average charges of $280,792.73 were significantly higher compared to the living donor transplant average charges of $223,529.35 (P <0.0001) (Fig. 2). Medicare payments were also significantly higher for cadaveric donor transplants ($118,099.35) compared to living donor transplants ($96,060.48) (P <0.0001) (Fig. 2).
Transplant stay charges were significantly higher for cadaveric versus living donation.
Transplant related service charges accrued during the inpatient transplant hospitalization were compared for patients selected as having primary Medicare coverage. These total transplant stay charges represent only the immediate posttransplant inpatient stay until discharge. Cadaveric charges ($79,730) were significantly higher (P <0.0001) than living donation ($69,547) when Medicare was selected as the primary payer (Fig. 3).
Transplant stay payments for inpatient hospital services demonstrate no significant difference between cadaveric and living donor transplantation.
Comparing the inpatient portion of the transplantation and related services for patients selected as having primary Medicare coverage, average transplant stay payments were $28,447 for living donation versus $28,483 for cadaveric donation (Fig. 3). There was no statistical difference between groups (P =0.858).
The limited number of donor organs continues to be a problem in all of solid organ transplantation. Increasing the donor pool with xenografts or prolonging graft survival via tolerance still has limitations. Increasing the donor pool in renal transplantation with living donation has proven to be successful for many years. Improved graft and patient survival for multiple reasons are certainly advantages for patients, as well as third-party payers and transplant centers. Only recently, however, have the cost implications of this new donor source been addressed.
It is not too surprising that the main source of cost savings of living compared to cadaveric donation is the inpatient hospital setting. Given the graft survival data and frequency of rejection episodes comparing living versus cadaveric grafts, one would predict higher costs for laboratory, immunosuppression, etc. The cost savings in the other categories may not have been so obvious. With continued research and statistical analysis, we hope to refine our understanding of the sources of the cost savings of living donation.
Using the criteria listed above in conjunction with the acquired data, profitability can be inferred by using the payment to charge ratios. From the comparison of transplant charges and payments for patients with primary Medicare coverage, one can see that charges are higher for cadaveric donation compared to living donation but the payments are nearly identical (Fig. 3). In fact, the ratio of transplant stay payments to transplant stay charges is 41% for living donor transplants compared to 36% for cadaveric donor transplants. Because we believe charges are more highly correlated with costs than payments are with costs, we infer a higher profitability for living donor transplants.
The current study focused on the posttransplant period only. There are many other components of cost, including pretransplantation associated costs, that have already been well described (15). Additionally, organ acquisition costs were not included because these data were not accessible. Although payments, charges, and true costs are related, the terminology should not be confused. Costs, for example, include direct and arbitrarily allocated indirect costs. Charges tend to be higher than costs, whereas payments may be higher or lower than true costs based on payment methodology and the services provided to each.
With increasing technology, improved record keeping data bases, and clearer definitions of costs, the possibilities for cost studies become endless. The results of this work and others will provide an understanding of the beneficiaries of the cost savings from living kidney donation and allow the development of effective financial policies to promote the expansion of living donation programs.
The current state of renal transplantation involves the economics of supply and demand. Solutions, including clinical and experimental advances in the setting of managed care, will likely involve economic consideration. The potential cost savings with living donation compared to cadaveric donation, under current Medicare reimbursement policy, have great implications at the local, state, and federal level. The savings associated with living donation not only benefit Medicare and transplant centers, but patients as well. These savings could be used to further promote the benefits of living donation. Future analysis will hopefully involve a greater exploration of the sources of those cost differences.
1. White MG. Economics of transplantation. In: Primer on Transplantation, 1997. Norman DJ, Suki WD. American Society of Transplant Physicians, Thorofare, NJ, p. 179.
2. Evans RW. Organ transplantation costs, insurance coverage, and reimbursement. In: Terasaki PI, ed. Clinical transplants, 1990. Los Angeles: UCLA Tissue Typing Laboratory, 1990; 343.
3. Evans RW. Third party payments and the cost of transplantation. J SC Med Assoc 1991; 87:345.
4. Evans RW, Kitzmann DJ. An economic analysis of kidney transplantation. Surg Clin North Am 1998; 78:149.
5. Eggers PW, Kucken LE. Cost issues in transplantation. Surg Clin North Am 1994; 74 (5):1259.
6. Eggers P. Comparison of treatment costs between dialysis and transplantation. Semin Nephrol 1992; 12:284.
7. Blommer TJ, Schanbacher B, Corry RJ. Transplant and dialysis: the cost/benefit question. Iowa Med 1984; 74:15.
8. Terasaki PI, Cecka JM, Gjertson DW, Takemoto S. High survival rates of kidney transplants from spousal and living unrelated donors. N Engl J Med 1995; 333:333.
9. Daar AS. Living-organ donation: time for a donor charter. In: Terasaki PI, Cecka JM, eds. Clinical Transplants 1994. UCLA, 1995; 376.
10. Karlberg I. Cost analysis of alternative treatments in end stage renal disease. Transplant Proc 1992; 24 (1):335.
11. U.S. Renal Data System. Researchers’ guide to the USRDS database. Bethesda, MD: National Institute of Health, National Institute of Diabetes and Digestive and Kidney Diseases, 1998.
12. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958; 53:457.
13. DiCiccio TJ, Efron B. Bootstrap confidence intervals. Stat Sci 1996; 11:189.
14. Efron B. Better bootstrap confidence intervals. J Am Stat Assoc 1987; 82:171.
15. Evans RW, Manninen DL, Dong FB. An economic analysis of liver transplantation. Gastroenterol Clin North Am 1993; 22 (2):451.