| 2008 JASN IMPACT FACTOR 7.505 | HOME AUTHOR INFO EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP | |||
| CURRENT ISSUE | ARCHIVES | JASN Express | ONLINE SUBMISSION | |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Clinical Transplantation |

,
,||
* Departments of Medicine,
Surgery,
Health Services Research, Management and Policy, University of Florida, Gainesville;
Florida Rehabilitation Outcomes Research Center and || Stroke Quality Enhancement Research Initiative, Veterans Affairs Health Services Research and Development/Rehabilitation Outcomes Research Center, Center of Excellence, North Florida/South Georgia Veterans Health System, Gainesville, Florida
Address correspondence to: Mr. Jesse Dylan Schold, Research Programs and Services, Division of Nephrology, Hypertension and Transplantation, University of Florida College of Medicine, PO Box 100224, Gainesville, Florida 32610-0224. Phone: 352-846-2692; Fax: 352-392-5465; E-mail: scholjd{at}medicine.ufl.edu
Received for publication May 17, 2005. Accepted for publication July 19, 2005.
| Abstract |
|---|
|
|
|---|
65 yr, odds ratio [OR] = 2.1, P < 0.01) relative to recipients aged 18 to 24 yr. African American and Asian recipients had a greater likelihood of receiving lower-quality organs relative to non-Hispanic Caucasians. Regional allocation networks were highly variable with regard to donor quality. Neither recipient gender (OR = 1.00, P = 0.81) nor patients primary diagnosis were associated with donor quality. Findings suggest that disparities in the quality of deceased donor kidneys to transplant recipients exist among certain patient groups that have previously documented access barriers. The extent to which these disparities are in line with broad policies of equity and potentially modifiable will have to be examined in the context of allocation policy. | Introduction |
|---|
|
|
|---|
When patients are placed on the deceased-donor waiting list, kidneys are allocated based on national policy as rigorously defined by the OPTN (12). This national kidney allocation policy accounts for regional factors, recipient panel reactive antibody (PRA) level, human leukocyte antigen (HLA) matching, waiting time on dialysis, blood type, and patient age (only adult versus pediatric). Additionally, there remains a certain degree of discretion by hospitals, Organ Procurement Organizations (OPO), and physicians concerning the appropriateness of transplanting donor organs to a particular recipient given the specific conditions of the potential transplant. Ultimately, patients and patient advocates must decide whether to accept a particular organ given their particular circumstances and the available information regarding the nature of the donated organ. To further add to the complexity of the transplant process, the characteristics of a potential donor organ are highly variable. Ideally, candidates benefit most from a living donor transplant. However, even among deceased-donor transplants there exists wide variability in the quality of the donated kidneys and consequently the expected outcomes of patient and graft survival. In fact, recent research reported a donor risk categorization (grades I to V) with a nearly three-fold adjusted risk for graft loss associated with the lowest quality (grade V) deceased-donor kidney relative to the highest quality (grade I) (13). In an attempt to streamline the allocation of organs of marginal quality, the United Network of Organ Sharing (UNOS) established the expanded donor criteria (ECD) in 2002, which dichotomously characterized kidneys at high-risk, defined as those kidneys that conveyed a 70% or greater increased risk of graft loss relative to standard kidneys (14).
The principal aim of our study was to assess the degree to which previously documented factors associated with access barriers to transplantation among ESRD patients exist with regard to access to high-quality, deceased-donor kidneys. A secondary aim was to assess the impact of both national and alternative allocation processes in contributing to any existing disparities and subsequent outcomes.
| Materials and Methods |
|---|
|
|
|---|
Measures
We categorized patient level data derived from applicable transplant forms by demographic characteristics. We recoded recipient age into the following age groups: 18 to 24, 25 to 34, 35 to 44, 45 to 54, 55 to 64, and
65 yr. Race/ethnicity was recoded into six categories: (1) Hispanic; (2) Caucasian, non-Hispanic; (3) African American, (4) Native American; (5) Asian/Pacific Islander; and (6) other (including multiracial, native Hawaiian or other Pacific Islander, Arab or other Middle Eastern, and Indian subcontinent). Indications of Hispanic ethnicity were used as a separate grouping regardless of race designation; however, 93.4% of the cohort indicating Hispanic ethnicity was also listed as Caucasian. Recipient PRA percentage was categorized into groups of 0, 1 to 10, 11 to 30, and
31. The recipients primary diagnosis was categorized into hypertension, glomerulonephritis, neoplasms, diabetes, congenital disorders, and other groupings. Pretransplant dialysis time was categorized as 0 to 6 mo, 7 to 24 mo, 25 to 36 mo, and
36 months. We used the 11 defined OPTN regions as a measure of the impact of geographic location as defined in the database. The states included in each region were as follows: Region 1 (CT, ME, MA, NH, RI), region 2 (DE, MD, NJ, PA, WV, Washington DC), region 3 (AL, AR, FL, GA, LA, MS, Puerto Rico), region 4 (OK, TX), region 5 (AZ, CA, NV, NM, UT), region 6 (AK, HI, ID, MT, OR, WA), region 7 (IL, MN, ND, SD, WI), region 8 (CO, IA, KS, MO, NE, WY), region 9 (NY, VT), region 10 (IN, MI, OH), region 11 (KY, NC, SC, TN, VA). Recipient primary insurance provider was categorized as: Medicare, Medicaid, Private, or other insurance (including the Department of Veterans Affairs, self-pay, and foreign sources). Utilization rates were calculated as the percentage of transplanted kidneys of all those recovered for transplant.
Outcome Measure
We used the Schold et al. (13) five-risk strata to categorize deceased-donor kidneys into quality groups (grades I to V). These quality grades incorporated the weighted effects of characteristics related to the quality of the donor organ and donor and recipient match based on a model for graft loss. The donor risk levels incorporated donor age, donor race, HLA (A, B, and DR) matching, cold ischemia time, donor and recipient cytomegalovirus status, donor cause of death, and donor history of diabetes and hypertension. This study reported progressive associations of donor grade with the risk for death-censored graft loss and overall graft loss. We also replicated models using standard ECD dichotomous stratification, which incorporates donor creatinine, donor history of hypertension, donor cause of death, and donor age as factors in defining high-risk organs utilized for transplantation.
Statistical Analyses
The dependent variable representing the five donor risk grades had five ordered levels (1 = lowest risk through 5 = highest risk). Therefore, we generated a multivariate, cumulative logit model to test the adjusted odds ratio (AOR) of receiving a lower-quality kidney relative to the defined reference groupings. Binary logistic models were utilized for the ECD designation as the dependent variable. Variables of interest included recipient age, recipient gender, recipient race/ethnic group, OPTN region, primary diagnosis, and primary insurance type. The model was additionally adjusted for waiting time on dialysis, recipient PRA level, and blood type. We also examined the interaction of recipient race (limited to non-Hispanic Caucasian and African American) and recipient age for the outcome of receiving a lower-grade kidney. In a separate adjusted Cox proportional hazard model, we examined the interaction of donor and recipient race for the outcome of overall graft loss.
2 tests were used to test the unadjusted independence of recipient characteristics with donor characteristics. Survival models were censored by the last individual follow-up variable available in database, with the last period through July 2003. Hypothesis tests and confidence intervals (CI) used 0.05 as the type I error level. All analyses were conducted using SAS v.9.1 (Cary, NC).
| Results |
|---|
|
|
|---|
65 yr (27.6%). The rate of ECD transplants in Asian recipients (18.1%, P = 0.02) and other race groups (20.0%, P = 0.01) were higher relative to non-Hispanic Caucasians (16.0%).
|
65 having a >five-fold higher likelihood of receiving a high-risk transplant relative to recipients 18 to 24 yr old. In this same model, as compared with non-Hispanic Caucasians, African Americans had 48% greater odds (AOR = 1.48, CI = 1.42, 1.55), Asians/Pacific Islanders had 76% greater odds (AOR = 1.76, CI = 1.61, 1.92), Hispanics had 19% greater odds (AOR = 1.19, CI = 1.12, 1.26), and other ethnic groups 47% greater odds (AOR = 1.47, CI = 1.26, 1.70) of receiving a lower-grade kidney. However, when using the ECD designation as the response variable, the magnitude of the association of donor risk and race/ethnicity was substantially reduced, but remained statistically significant. Recipient gender was not significantly associated with donor quality in the adjusted model, male recipients (AOR = 1.00, CI = 0.96, 1.03) relative to females. Recipient primary insurance type was significantly associated with donor quality using the five donor grades, but showed no significant association with receipt of an ECD transplant.
|
|
|
| Discussion |
|---|
|
|
|---|
The disparity in donor organ quality by recipient age suggests that allocation processes beyond those outlined in national policy play a significant role in determining donor and recipient match. As age is not a factor in allocation for adult patients, the observed disparities are likely a function of noncodified processes. These alternative processes potentially include patient consent as well as physician and center selection. One explanation for our observations is that a tacit policy of steering lower-quality organs to older age recipients is practiced at a local or center level. It is possible that older patients, who are sicker on average, are also more likely to accept lower quality organs as an alternative to remaining on dialysis. The progressive association of lower donor quality with older recipients appeared to be consistent in all donor grades, with the possible exception of the lowest-risk organs (grade I). As more than one third of the grade I organs were 6-antigenmatched transplants, this may suggest that this policy deters some of the local distribution patterns. This association was even stronger when utilizing the ECD designation as the effects appear strongest among the highest-risk organs. Our results were also congruent when restricting the study population to recipients that were not sensitized, constituting a more homogenous cohort that was not as limited by donor cross-match. The ECD policy, implemented in late 2002, voluntarily consents patients to receive high-risk kidneys, and was considered to result in shorter waiting times for these transplants. Our findings indicated that older recipients were more likely to receive high-risk organs before implementation of this policy. How this policy will affect future allocation patterns may depend on the manner by which candidates are consented to receive these higher risk organs. However, as donor risk represents a wide continuum of quality beyond the ECD designation, alternative processes will likely continue to influence allocation. As the association of donor quality and recipient age is relatively strong and progressive, this may suggest that there is some degree of matching higher-quality kidneys with younger patients with longer expected survival. In fact, recent research has suggested that there may be significant utilitarian and economic benefits from this mechanism of allocation (16). Although this relationship may be justified in certain circumstances, the question remains whether it would be useful to standardize this process more rigorously (allocating kidneys based on age or survival expectancy) rather than leave this to variable application and other subjective criteria that could potentially detrimentally affect those groups with access barriers. Additional factors associated with patient access to higher quality of care among this population not ascertainable in this analysis may include education and income levels, and those candidates more proactive and involved in the transplant process. In addition, physician perceptions of patients social support network or patients proclivity to be noncompliant with medication regimens may also contribute to observed disparities.
A certain degree of variation is probably to be expected across the eleven transplant regions due to demographic variations in the population as well as differences in the medical communities. The risk level for lower-quality donations also appeared to correlate with the regions with extended waiting times. Region 9 had an elevated risk of lower-quality donation and also had the longest waiting time for transplant from a 2004 OPTN report (17). According to this report, among patients with type O blood who were wait-listed in 1997 and 1998, the median waiting time in region 9 for a transplant was 7.1 yr, whereas no other region reported a waiting time >5.1 yr. Conversely, region 8, which represented the lowest donor risk among all regions, had the shortest waiting time for transplant at 2.6 yr. As time on dialysis confers a heightened risk for patient death and future graft loss, it is possible that patients with longer waiting times may be less selective regarding organ quality (18). However, as regions with lower-quality transplants also had higher discard rates, it may not simply be heightened selectivity that drives the variability in regional quality. This is in no way an indictment of procurement or transplant practice in particular regions; in fact, significant variation likely exists at a much more granular level within each region. However, the question arises whether the regional networks could be more optimally aligned to balance donor quality across geographic areas. Potential recipients who are cognizant of these issues may gain significant advantages by listing (or multiple-listing) in areas with lower waiting times and higher donor quality relative to those without this knowledge. Clearly this is a complex issue, and beyond the scope of this analysis, but it is important to recognize donor quality when considering regional alignment, equity, and allocation policy.
Results utilizing the five-level donor-risk strata suggest a relatively large association between donor quality and race/ethnic group; however, when utilizing the ECD criteria, which incorporates neither impact of HLA-matching nor donor race, the magnitude of the effect was substantially decreased. HLA-matching has previously been documented to disproportionately negatively impact outcomes in African Americans (5,19). The lower availability of potential HLA-matched organs for minority groups has been addressed previously, and recently the elimination of allocation points for the B antigen (other than for 6-antigen matches) is thought to ameliorate some degree of racial disparities (20,21). In addition, there have been considerable efforts to increase donation rates among minorities, which would also likely help rectify quality disparities. One implication of our results could be that further de-emphasis or elimination of HLA-matching in allocation policy would reduce the disparity of donor quality by race. As the benefit of matched antigens to the recipient is partially driven by decreased waiting time through the allocation point system, the effect of reducing its effect should result in more equitable distribution. In addition, this policy change could have the secondary effect of allowing for better management of the waiting list given more fixed knowledge concerning the priority of candidates and more transparent patient information regarding their expected time to transplant. However, the dynamic relationship between overall utility of resources and equitable distribution are often competing interests, and as HLA-matching has always been associated with increased graft survival, such amendments must be balanced with potential deleterious effects. Another contribution to the disparities in quality derives from the allocation of deceased-donor transplants from African Americans. African American kidneys have previously been reported as a risk factor for graft loss (22,23). Findings from this study demonstrated that this risk is present among African American recipients, yet African American donations are disproportionately allocated to African American recipients. This association may be affected by geographic factors as well as HLA-matching. Whether realigned transplant regions would ameliorate these disparities is another important consideration for allocation policy. Our analysis does not imply that outcomes among racial and ethnic groups could be equated with a shift in allocation policy alone, as our results indicated that outcomes among African American recipients were diminished relative to Caucasians even with the highest donor quality.
Equitable access to healthcare is a pervasive issue in our society, with economic, ethical, and sociological implications. The ESRD population in the US has grown substantially over the past 20 yr and access to appropriate health care services for this population is a growing societal concern. There have been limited improvements in deceased-donor transplant rates over the past decade despite valiant efforts to bridge the growing chasm between the need for transplantation and available organs. One of the implications of this disparity is the growing value of the available donations. While alternative allocation strategies likely exist that enhance the overall graft survival of this population, the implications for particular subgroups must always be considered. Allocation policy and donor quality are complex issues that elicit highly asymmetric knowledge between patients and caregivers. In this sense, much of the responsibility to ensure equitable and standardized processes rests with the transplant community. One of the challenging paradoxes concerning allocation policy is that for individual patients there is significant incentive to acquire the highest-quality donor organs, but from a collective utilitarian perspective, the use of all donatious (including higher-risk organs) is beneficial to this population as a whole (24). As we have argued, transplantation represents a wide spectrum of access potential, and policies that govern allocation should consider the aspect of donor quality with all equitable considerations.
| Acknowledgments |
|---|
This material is the result of work supported with resources and the use of the facilities at the Department of Veterans Affairs Health Services Research and Development/Rehabilitation Research and Development/Rehabilitation Outcomes Research Center, North Florida/South Georgia Veterans Health System.
A portion of this material was presented as an oral abstract at the annual American Transplant Congress in Seattle, WA, May 21 to 25, 2005.
| Footnotes |
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
J. Schold, T. R. Srinivas, A. R. Sehgal, and H.-U. Meier-Kriesche Half of Kidney Transplant Candidates Who Are Older than 60 Years Now Placed on the Waiting List Will Die before Receiving a Deceased-Donor Transplant Clin. J. Am. Soc. Nephrol., July 1, 2009; 4(7): 1239 - 1245. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. J. Gordon and J. C. Caicedo Ethnic advantages in kidney transplant outcomes: the Hispanic Paradox at work? Nephrol. Dial. Transplant., April 1, 2009; 24(4): 1103 - 1109. [Full Text] [PDF] |
||||
![]() |
S. Ramirez Race and Kidney Disease Outcomes: Genes or Environment? J. Am. Soc. Nephrol., December 1, 2005; 16(12): 3461 - 3463. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
HOME
CURRENT ISSUE
ARCHIVES
JASN Express
ONLINE SUBMISSION
AUTHOR INFO
EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP |