|
Consequences of eliminating HLA-B in deceased donor kidney allocation to increase minority transplantation |
Mutinga N, Brennan D C, Schnitzler M A |
|
|
Record Status This is a critical abstract of an economic evaluation that meets the criteria for inclusion on NHS EED. Each abstract contains a brief summary of the methods, the results and conclusions followed by a detailed critical assessment on the reliability of the study and the conclusions drawn. Health technology The elimination of HLA-B matching in deceased donor kidney allocation was examined.
Study population The study population comprised a hypothetical cohort of patients with ESRD who were waiting for a kidney transplant. The four ethnic groups considered were Caucasian, African American, Asian and others.
Setting The setting was a hospital. The economic study was carried out in the USA.
Dates to which data relate The effectiveness and resource use data were derived from a database gathering data from 1995 to 1999. Further studies published in 1996 and 2004 were also used. The price year was 2000.
Source of effectiveness data The effectiveness evidence was derived from a synthesis of published studies.
Modelling A Markov model was constructed to assess the long-term expected outcomes and costs of transplantation including and excluding HLA-B matching in the allocation scheme. Following transplantation, patients ended the first time period in one of three states (function of the transplant, return to dialysis, or death). Patients on dialysis ended each period in one of two states (survival on dialysis or death). The probability of ending a period in a particular state depended on the previous state, the period, and whether the patients had received a "well-matched" kidney or not. Re-transplantation was not modelled, because of the small number of patients receiving it. The time horizon of the model was 20 years, but the analysis was also replicated using a timeframe of 15 years. The cycle length was one year. A schematic representation of the model was provided.
Outcomes assessed in the review The outcomes estimated from the literature were:
the rate of deceased donor graft survival at 1, 2, 3 and 5 years (based on observed multivariate adjusted survival with HLA-B);
the expected survival without HLA-B at 1, 2, 3 and 5 years (and relative hazard ratio);
the rates of death and wait-list survival;
the impact of elimination of HLA-B (number of organs shifted from Caucasian to minorities); and
the utility scores.
Study designs and other criteria for inclusion in the review It appears that the authors have identified the primary studies selectively and not by undertaking a review of the literature. Most of the clinical data came from a published database, the United States Renal Data System (USRDS), which contained information on patients transplanted or awaiting transplantation between 1995 and 1999. Details on the source of the utility values were not reported.
Sources searched to identify primary studies Criteria used to ensure the validity of primary studies Methods used to judge relevance and validity, and for extracting data Not stated. However, a multivariate Cox regression analysis was used to calculate estimates of graft survival at 1, 2, 3 and 5 years, post-transplant and graft survival at year 4, and post-transplant given survival to year 2. Further, the survival of wait-listed patients on dialysis was adjusted for mean patient age, race, gender, cause of ESRD and blood type.
Number of primary studies included Three primary studies provided evidence.
Methods of combining primary studies Not relevant since the primary data were not combined.
Investigation of differences between primary studies Results of the review The rates of deceased donor graft survival with HLA-B at year 1 were 88.0% for Caucasians, 87.2% for African Americans, 97.2% for Asians and 90.6% for others.
The rates of deceased donor graft survival with HLA-B at year 2 were 84.3% for Caucasians, 81.1% for African Americans, 90.0% for Asians and 87.8% for others.
The rates of deceased donor graft survival with HLA-B at year 3 were 80.4% for Caucasians, 75.3% for African Americans, 87.4% for Asians and 84.9% for others.
The rates of deceased donor graft survival with HLA-B at year 5 were 72.0% for Caucasians, 65.9% for African Americans, 82.9% for Asians and 80.2% for others.
The long-term graft loss rates (calculated from year 3 to year 5) were 5.4% for Caucasians, 6.4% for African Americans, 2.6% for Asians and 2.8% for others.
For expected survival without HLA-B, the relative hazard ratios were 1.02 for all ethnic groups.
The rates of deceased donor graft survival without HLA-B at year 1 were 87.8% for Caucasians, 86.9% for African Americans, 92.6% for Asians and 90.4% for others.
The rates of deceased donor graft survival without HLA-B at year 2 were 84.0% for Caucasians, 80.7% for African Americans, 89.8% for Asians and 87.6% for others.
The rates of deceased donor graft survival without HLA-B at year 3 were80.0% for Caucasians, 74.8% for African Americans, 87.1% for Asians and 84.6% for others.
The rates of deceased donor graft survival without HLA-B at year 5 were 71.4% for Caucasians, 65.2% for African Americans, 82.6% for Asians and 79.8% for others.
The long-term graft survival rates at year 5 were 5.5% for Caucasians, 6.6% for African Americans, 2.7% for Asians and 2.9% for others.
The rates of death with a functioning graft were 44.5% for Caucasians, 34.1% for African Americans, 39.7% for Asians and 34.4% for others.
The rates of death risk after graft loss were 20.5% for Caucasians, 14.7% for African Americans, 13.9% for Asians, and 13.4% for others within one year. The rates through year 2 were 27.5% for Caucasians, 19.9% for African Americans, 19.0% for Asians and 17.8%. The rates after one year were 7.0% (Caucasians), 5.2% (African Americans), 5.1% (Asians) and 4.7% (others), respectively.
In terms of the waiting-list survival, the rates of 4-year patient survival on the wait-list given 2-year survival were 83.0% for Caucasians, 89.4% for African Americans, 92.6% for Asians and 87.3% for others. The rates of death avoided by transplant (average death rate between 2 and 4 years of wait) were 8.9% for Caucasians, 5.4% for African Americans, 3.8% for Asians and 6.6% for others.
With the elimination of HLA-B matching, 166 kidneys would shift from Caucasian to minorities. In particular, African Americans would receive 138 additional transplants (2,292 total), Asians 21 additional transplants (357 total), and other minorities 7 additional transplants (139 total). The utility scores for dialysis were 0.6. The utility benefit of transplantation was 0.26.
Measure of benefits used in the economic analysis The summary benefit measure was the quality-adjusted life-years (QALYs). These were calculated by combining expected survival with quality of life data. An annual discount rate of 5% was applied.
Direct costs The cost analysis was performed from the perspective of the payer. It included all medical costs related with transplantation, such as kidney acquisition, hospitalisations and dialysis. A detailed breakdown of the cost items was not reported, as the costs were estimated from average Medicare reimbursement rates. A 20% co-payment was added to Medicare payments because patients or secondary insurance co-payments were not recorded in the database. Kidney acquisition was estimated from the Center for Medicare Studies. The estimation of resource use data was not described clearly. Discounting was relevant since a 20-year time horizon was used in the model, and an annual rate of 5% was applied. The price year was 2000. Costs estimated in previous years were inflated to 2000 values using the medical component of the Consumer Price Index.
Statistical analysis of costs Medicare costs were calculated using multivariate regression, with adjustments for mean patient age, race, gender, degree of immunologic sensitisation, cause of ESRD, insulin dependence, length of time on dialysis, and the number of HLA mismatches.
Indirect Costs The indirect costs were not included in the economic analysis.
Sensitivity analysis A Monte Carlo sensitivity analysis was undertaken using 50,000 iterations and published ranges of values. Several univariate sensitivity analyses were performed to assess the robustness of the cost-utility ratios to variations in the utility benefits of transplant, discount rate, and the number of kidneys re-distributed by the elimination of HLA-B matching.
Estimated benefits used in the economic analysis The estimated QALYs were:
with HLA-B matching, 8.0 for Caucasians, 7.9 for African Americans, 9.4 for Asians and 9.3 for others;
without HLA-B matching, 7.9 for Caucasians, 7.8 for African-Americans, 9.4 for Asians and 9.2 for others; and
with dialysis, 4.1 for Caucasians, 5.1 for African Americans, 5.8 for Asians and 4.8 for others.
Cost results The estimated costs were:
with HLA-B matching, $374,842 for Caucasians, $420,827 for African Americans, $317,113 for Asians and $419,842 for others;
without HLA-B matching, $375,403 for Caucasians, $422,117 for African Americans, $317,991 for Asians and $419,998 for others; and
with dialysis, $399,361 for Caucasians, $496,786 for African Americans, $484,106 for Asians and $409,087 for others.
Synthesis of costs and benefits Average cost-utility ratios were calculated to combine the costs and QALYs of the alternative strategies. Over a 20-year timeframe, the average cost per QALY was:
with HLA-B matching, $46,863 in Caucasians, $53,417 in African Americans, $33,734 in Asians and $45,285 in others;
without HLA-B matching, $47,265 in Caucasians, $53,975 in African Americans, $33,953 in Asians and $45,485 in others;
with dialysis, $97,911 in Caucasians, $96,568 in African Americans, $83,237 in Asians and $85,981 in others.
Most interesting results were obtained when the figures were aggregated.
Over a 20-year time horizon, the elimination of HLA-B matching (and consequently a shift of 166 kidneys from Caucasians to minorities) resulted in a loss of $7.0 million and 697 QALYs to Caucasians. However, it led to an increase of 243 QALYs and $7.5 million in cost-savings in African Americans, 64 QALYs and $3.2 millions cost-savings in Asians, and 11 QALYs at an increased cost of $160,000 in other minorities. This translated into net savings of $3.5 million, but at the expense of a net loss of 649 QALYs. The expected cost-savings per lost QALY were $5,400. A comparable result was achieved when a 15-year timeframe was used (the cost-savings per lost QALY were $465). Thus, the elimination of HLA-B matching would lead to cost-savings, loss of QALYs and redistribution of QALYs among ethnicities.
Excluding the harms and benefits from the redistribution of organs, the elimination of HLA-B matching from the allocation scheme reduced the total QALYs and increased the total costs in each racial group. In total, the elimination of HLA-B matching contributed 348 QALYs to the total lost from this change in policy and reduced the cost-savings by $2.7 million. Thus, the elimination of HLA-B matching with direct allocation saved $21,000 per QALY lost.
The Markov model showed also that 284 additional Caucasian donors would have reduced the total costs by $15.7 million and increased the QALYs by 1,487, while 201 additional African-American donors would have reduced the total costs by $13.3 million and increased the QALYs by 997. Thus, increasing donation would be much more cost-effective than eliminating HLA-B matching. The sensitivity analysis confirmed the robustness of the base-case estimates. Increasing organ donation increased QALYs and reduced costs. On the other hand, the elimination of HLA-B matching would reduce QALYs and might increase costs. The univariate sensitivity analysis showed that increasing (reducing) the utility benefit of transplant over dialysis by 30% increased (decreased) the number of QALYs lost to 712 (586) and reduced (increased) the costs saved per lost QALY to $4,900 ($6,000). The impact of variations in the other base-case inputs was also reported.
Authors' conclusions A policy of eliminating HLA-B matching in deceased donor kidney allocation was expected to reduce the total quality-adjusted life-years (QALYs), although it saved resources relevant to the third-party payer. In particular, the QALYs lost by the Caucasians would be higher than the QALYs gained by African Americans, Asians and other minorities.
CRD COMMENTARY - Selection of comparators The authors provided a justification for the choice of the comparators, the selection of which was appropriate for the study question. You should decide whether they are valid comparators in your own setting.
Validity of estimate of measure of effectiveness The bulk of the effectiveness evidence came from a large administrative database, supplemented with clinical data derived from other published studies. Limited information on the characteristics of this database was reported. However, the use of administrative data permits the use of large samples of patients, but at the expense of data that are generally difficult to disaggregate. Further, details of the other sources of evidence were not given. Thus, it was difficult to assess the validity of the other sources. The utility weights came from the literature, but the source of such values was not reported. Owing to uncertainty in the clinical estimates, some sensitivity analyses were undertaken.
Validity of estimate of measure of benefit The benefit measure used in the analysis was appropriate as QALYs capture the impact of the interventions on the most relevant dimensions of care (i.e. survival and quality of life). Further, QALYs are comparable with the benefits of other health care interventions. Discounting was applied. Limited information on the source of the utility weights was provided.
Validity of estimate of costs The included costs were consistent with the perspective adopted in the study. The costs were presented as macro-categories and a detailed breakdown of the cost items was not provided. This reduces the possibility of replicating the analysis in other settings. Moreover, the unit costs were not presented. The source of the costs, which was provided, was consistent with the perspective of the analysis. However, limited information on resource consumption was given. Statistical analyses of the costs were not performed and the cost estimates were quite specific to the study setting. The impact of different discount rates was investigated in the sensitivity analysis. The price year was reported, which aids reflation exercises in other settings.
Other issues The authors did not make extensive comparisons of their findings with those from published studies, although they did report the success of some pilot organ donor programmes in Europe. The issue of the generalisability of the study results to other settings was not explicitly addressed and few sensitivity analyses were performed. Thus, the external validity of the study was limited. The analysis referred to patients awaiting kidney transplantation and this was reflected in the authors' conclusions. Some limitations of the analysis were also noted. For example, the use of a large administrative database as the main source of clinical data and the ideal scenarios considered within the modelling framework, which might not reflect real-world conditions. The results of the analysis were reported selectively. Finally, the authors calculated the cost per QALY lost rather than the cost per QALY gained, which is generally used. The authors acknowledged that other important issues, in addition to the cost-effectiveness of programmes, may arise, in particular those associated with ethical aspects.
Implications of the study The authors suggested that "future efforts to assist minority patients awaiting transplantation may be better directed to organ donation than adjustments to existing organ allocation systems".
Source of funding Supported in part by grants from the National Institute of Diabetes, Digestive, and Kidney Diseases (k25-dk-02916-03, k24-dk-002886-02 and T32-DK-071126-01).
Bibliographic details Mutinga N, Brennan D C, Schnitzler M A. Consequences of eliminating HLA-B in deceased donor kidney allocation to increase minority transplantation. American Journal of Transplantation 2005; 5(5): 1090-1098 Other publications of related interest Whitling JF, Kiberd B, Kalo Z, et al. Cost-effectiveness of organ donation: evaluating investment into donor action and other donor initiatives. Am J Transplant 2004;4:569-73.
Roberts JP, Wolfe RA, Bragg-Gresham JL, et al. Effect of changing the priority for HLA matching on the rates and outcomes of kidney transplantation in minority groups. N Engl J Med 2004;350:545-51
Indexing Status Subject indexing assigned by NLM MeSH African Americans; Cadaver; Cost-Benefit Analysis; Ethnic Groups; European Continental Ancestry Group; HLA-B Antigens /metabolism; Histocompatibility Testing; Humans; Kidney Transplantation /economics /methods; Markov Chains; Monte Carlo Method; Multivariate Analysis; Outcome Assessment (Health Care); Proportional Hazards Models; Quality of Life; Quality-Adjusted Life Years; Registries; Resource Allocation; Sensitivity and Specificity; Tissue Donors; Tissue and Organ Procurement; Treatment Outcome; United States AccessionNumber 22005000889 Date bibliographic record published 28/02/2006 Date abstract record published 28/02/2006 |
|
|
|