Kidney transplantation is considered the best long-term treatment for patients with end-stage kidney disease, yet less than 20% of incident patients are wait-listed for transplant within 1 year of starting dialysis.

Transplantation requires referral to a transplant center, followed by an evaluation process and placement on a waitlist. Prior studies have demonstrated significant regional variation in referral practices and, by extension, variation in subsequent steps of the transplant process.

Compounding this issue, there are also significant racial differences in the incidence and prevalence of ESKD, as well as in the modes of renal replacement therapy utilized.

Although Black individuals have higher rates of ESKD, white individuals are overrepresented on kidney waitlists and receive a disproportionate number of transplants. Prior studies have identified racial disparities in transplant rates between underrepresented groups and white individuals. The reasons for racial disparities in transplant rates may include some of the following:

differences in referral rates;
completion of the transplant evaluation;
perceived medical suitability;
insurance coverage and access to adequate medication coverage; and
preferences for ESKD treatment.

However, at present it is unclear at which step (or steps) in the process racial disparities emerge, whether it occurs at referral for evaluation, the waitlist process or transplantation.

Transplant referrals

Despite the importance of transplant referral in conferring access to kidney transplantation, there is no national database for tracking these referrals, making it difficult to study on a national level. Assessment of referral rates has gained even greater importance since CMS added the percentage of prevalent patients on dialysis listed for transplant as a Quality Incentive Program metric in 2018.

Similarly, the ESRD Treatment Choices Model includes the percentage of ESKD patients starting on dialysis who are wait-listed or transplanted, with a living donor kidney as a central outcome metric. However, in the absence of a national database for tracking transplant referrals, data derived from the referring dialysis clinics remain the only currently available source of information on this key step in the transplant process.

Knowledge gaps

In a study, we sought to address important knowledge gaps regarding the associations of patient race with kidney transplant referral rates and to explore how differences in referral may be reflected in later steps in the transplant process. A single data source from a national dialysis organization was utilized to compare rates of kidney transplant referral, waitlist and transplantation between incident white and Black patients on dialysis. This retrospective study used electronic health records and county-level indices of socioeconomic deprivation at the time of dialysis initiation at the dialysis organization.

Patients included were those who, between July 1, 2015, and June 30, 2018, were aged 18 to 80 years, were incident to dialysis and began care at the dialysis organization within 30 days of their first dialysis treatment.

Patients with prior transplant and those who had undergone transplant evaluation or listing prior to starting dialysis were excluded. Only Black patients and white patients were included in the study, as data for other groups were sparse.

Patients were followed from date of first eligibility until March 31, 2019, or censored for death, transfer, withdrawal from dialysis, renal recovery, transplantation or loss to follow-up. Outcomes were assessed across exposure categories using Cox proportional hazards models adjusted for the following indicators:

age, sex and BMI;
etiology of ESKD;
diabetes, congestive heart failure (CHF) and coronary artery disease (CAD);
amputation, Charlson comorbidity index, Organ Procurement and Transplant Network (OPTN) region; and
quartiles of the following county-level sociodemographic indicators: high school graduation rate, unemployment level, social associations, severe housing problems and poverty.

Primary outcomes of interest were threefold, each reflecting the successful completion of a discrete stage in the transplant process conditional upon having completed the prior stage: time from dialysis initiation to referral for transplant evaluation (among all patients initiating dialysis); time from referral to placement on the waitlist (among referred patients) and time from placement on the waitlist to receipt of a transplant (among waitlisted patients).

Secondary outcomes were twofold, reflecting the marginal likelihood of completing a downstream stage in the process, considering all eligible patients who began dialysis: time from dialysis initiation to wait-listing and time from dialysis initiation to receipt of a kidney transplant.

Study results

A total of 60,229 patients (61%) of self-identified as white and 23,499 patients (39%) as Black were included in the study (see Table). Black patients were, on average, younger than white patients; more likely to be female and have hypertension as the etiology of ESKD; tended toward lower Charlson comorbidity index scores and were more likely to have Medicaid insurance.

Black patients were also more likely to reside in ZIP codes with lower high school graduation rates, more severe housing problems and more poverty compared with white patients. These imbalances were generally preserved among the subgroups of patients who were subsequently referred and listed for renal transplant, respectively, and were adjusted for in subsequent analyses.

Among white patients initiating dialysis, the transplant referral rate was 0.28 events per patient year (95% CI, 0.28-0.29), whereas among Black patients, the rate was 0.43 (95% CI, 0.42- 0.44) events per patient year. Following adjustment for imbalanced characteristics, Black race was associated with a 22% higher likelihood of referral than white race (hazard ratio = 1.22; 95% CI, 1.19- 1.26).

Among the 20,087 patients from the initial study cohort who were referred for transplant during follow-up, 2,755 white patients and 2,454 Black patients were subsequently listed for transplant. The listing rate was 0.22 events per patient year (95% CI, 0.21-0.23) among white patients compared with 0.19 events per patient year (95% CI, 0.18-0.19) among Black patients. In multivariable analyses, Black race was associated with an 18% lower likelihood of transplant listing compared with white race (HR = 0.82; 95% CI, 0.78-0.88). Considering all patients who initiated dialysis, irrespective of referral status, Black race tended to be associated with a lower likelihood of transplant listing (HR = 0.94; 95% CI, 0.89-1).

Transplants

Among the 5,209 patients from the initial study cohort who were listed for transplant, 598 white patients and 277 Black patients received a transplant. The transplant rate was 0.12 events per patient year (95% CI, 0.11-0.13) among white patients and 0.06 events per patient year (95% CI, 0.05-0.06) among Black patients. Following adjustment, Black race was associated with a 53% lower likelihood of transplant compared with white race (HR = 0.47; 95% CI, 0.40-0.55). Considering all patients who initiated dialysis, regardless of referral or listing status, Black race was associated with a 55% lower likelihood of transplant receipt (HR = 0.45; 95% CI, 0.39-0.53).

Sensitivity analyses

Initial dialysis modality/vascular access type and primary insurance type might influence the likelihood of transplant referral, waitlist or receipt of transplant. In a sensitivity analysis, the relationships between self-reported race and each outcome were evaluated by adjusting for each of these two additional factors individually, or both simultaneously, while also controlling for the demographic and socioeconomic factors considered in the main analysis. These further adjustments did not substantially alter the estimated association of race with any outcome examined.

Associations between race and transplant listing may vary by OPTN region. When considering transplant listing among referred patients, inclusion of the interaction between OPTN region and race significantly improved model fit (likelihood-ratio test, P < .05). In most OPTN regions, the point estimate for Black race (vs. white race) was below 0.90 (regions 1, 2, 4, 5, 7, 8, 9, 10 and 11), although the trend only achieved nominal statistical significance in regions 2 and 7. The point estimate of the HR for transplant listing conditional upon referral was not greater than 1 in any OPTN region.

Discussion

In this study, Black patients were more likely to be referred for kidney transplantation but less likely to be wait-listed or to receive a transplant compared with white patients, suggesting potential disparities in later steps of the transplant process. Although Black patients are more likely to have B or O blood type, possibly resulting in longer wait times, we were unable to adjust for blood type due to lack of available data. That is a limitation of this study.

This study builds upon prior work regarding racial disparities in transplant access and raises the question of the location of the barrier to evaluation procedures and the transplantation process for these patients. Previous studies have suggested that, despite higher referral rates, Black patients are less likely to initiate and complete the evaluation process. It is also possible that transplant centers are less likely to deem Black patients as suitable candidates for transplant for medical or social reasons.

However, some evidence suggests that differences in medical eligibility alone do not explain the racial disparity. It is also possible that differences in the ability of transplant centers to contact a patient upon organ availability, or differential willingness to receive an organ from an extended-criteria donor, may contribute to the observed disparities in transplant receipt.

Navigating the transplant evaluation process is time-consuming and complex; it often requires multiple modalities of testing and referral to specialists. Additionally, the current finding suggests that efforts at increasing patient education about transplantation, while important, will likely not be sufficient to overcome racial disparities in the transplant process. It is essential to understand the relative impact of race on each of the additional steps, from the dialysis unit referral to placement on the waitlist, to rectify observed disparities in transplant access; additional study in this area is needed.

Unfortunately, placement on the waitlist does not guarantee access to transplantation. Modifications to the kidney allocation system to reduce emphasis on human leukocyte antigen matching and crediting wait time back to the start of dialysis have helped to narrow the access gap but have not eliminated it. Black patients with ESKD are still less likely to be preemptively wait-listed or receive a deceased donor transplant; they are also less likely to receive a living-donor transplant. Collectively, these data indicate that while understanding access to the kidney transplant waitlist is important, additional barriers to transplantation for Black patients remain. Addressing racial disparities in transplant wait-listing and receipt will require a focused examination of the patient evaluation and waitlisting process to identify opportunities for improvements in the equitable delivery of care.

For more information:
Bryan Becker, MD, MMM, FACP, CPE, is president of the Acclaim Multispecialty Group of the JPS Health Network, Fort Worth, Texas. He can be reached at bbecker@jpshealth.org.
Steven M. Brunelli, MD, MSCE, is vice president and medical director for health analytics and insights at DaVita Clinical Research, Minneapolis, Minnesota. He can be reached at steven.brunelli@davita.com.
Carey Colson, MBA, is a senior data research analyst at DaVita Inc. and can be reached at carey.colson@davita.com.
Jeffery Giullian, MD, is the chief medical officer at DaVita Inc. and is a member of the Editorial Advisory Board for Nephrology News & Issues. He can be reached at jeffrey.giullian@davita.com.
Deirdre Sawinski, MD, is an assistant professor of medicine at Weill Cornell Medical College and assistant attending physician at New York-Presbyterian Hospital. She can be reached at dls7001@med.cornell.edu.
Francesca Tentori, MD, is the vice president of outcomes research and patient empowerment at DaVita Clinical Research and can be reached at francesca.tentori@davita.com.

Published by:

Sources/Disclosures

Collapse

Disclosures: Becker, Brunelli, Colson, Giullian, Sawinski and Tentori report no relevant financial disclosures.

Read more

Racial disparity exists in the waitlist and transplant processes for Black patients