Interaction between socioeconomic deprivation and ethnicity for likelihood of receiving living-donor kidney transplantation

For many patients living with end-stage kidney failure, receiving a kidney from a living donor is the best treatment option in terms of survival and quality of life [1]. Despite this being widely acknowledged, rates of living-donor kidney transplantation (LDKT) have stagnated globally, and many barriers have been identified to explain this phenomenon [2]. Two of the most identified barriers to receiving a living-donor kidney are non-White ethnicity and socioeconomic deprivation [3‐10]. However, the interaction between these two factors remains poorly understood, and there are conflicting reports in the literature on this issue. For example, Gill and colleagues have observed disparity in the rates of living-donor kidney donation among African-Americans in the United States (US) when comparing higher versus lower income quintiles, suggesting lower rates for African-Americans in the lowest income group [4]. By contrast, Udayaraj and colleagues did not find any significant interaction between socioeconomic status and ethnicity in their population-cohort study in the United Kingdom (UK), suggesting that the effect of socioeconomic deprivation on the uptake of LDKT was similar across all ethnic groups [5].

For transplant centers that encompass large cohorts of both non-White demographics and areas of socioeconomic deprivation, it is important to clearly ascertain the interaction between these two variables, to ensure resources to encourage LDKT are targeted appropriately. Therefore, the aim of this study was to explore the proportions of living versus deceased donation, and explore the possible interaction between ethnicity and socioeconomic deprivation on LDKT, in a large single-center analysis encompassing an ethnically and socioeconomically diverse region of England.

In this retrospective single-center analysis, we have demonstrated that individuals who are non-White or live in socioeconomically deprived areas have a reduced likelihood of receiving a kidney transplant from a living-donor. Further analysis identified interplay between socioeconomic deprivation and ethnicity such that, whilst individuals of non-White ethnicity had a considerably lower chance of LDKT than their White counterparts in areas with high levels of deprivation, no such effect of ethnicity was observed in areas of low deprivation. Our study confirms the importance of ethnicity and socioeconomic deprivation as barriers to receiving LDKT, and introduces the important interaction between the two. While this study highlights the importance of encouraging discussions of the benefits of LDKT to individuals who live in socioeconomically deprived areas, it suggests appropriate tailoring of information is required for ethnic minority communities living in deprived areas, in order to achieve equity.

Non-White ethnicity and socioeconomic deprivation have long been associated with access to transplantation. For example, in a study of patients starting renal replacement therapy in the UK, Udayaraj and colleagues found those in the most deprived quintile (hazard ratio [HR]: 0.60, 95% CI: 0.54–0.68, p < 0.001) and of non-White ethnicity (HR = 0.89 for Black, 0.91 for South Asian) to be significantly less likely to waitlisted for a deceased donor kidney transplant [16]. Where patients go on to receive a transplant, our study suggests that both ethnicity and socioeconomic deprivation may influence the type of organ received. This finding is consistent with a population cohort analysis of registry data from the US performed by Reed and colleagues, who found a composite index of health and socioeconomic status factors to be negatively associated with LDKT, rates of which were 7.3 percentage points lower (95% CI − 12.2 to − 2.3, p = 0.004) in areas with greater burden of medical co-morbidity and more socioeconomic deprivation [17]. The study also observed lower rates of LDKT among transplant centers with higher prevalence of ethnic minorities, with rates 7.1 percentage points lower (95% CI − 11.8 to − 2.3, p = 0.004) [17]. However, the study did not assess the potential interaction between these two factors.

A recent study that did assess the interaction between socioeconomic status and ethnicity was a multi-center, cross-sectional study of a US cohort by Killian and colleagues. They quantified social deprivation using a social vulnerability index, and found both this and non-White ethnicity to be independently associated with a lower likelihood of LDKT [18]. However, they additional assessed the interaction between these factors and, like our study, found this effect to be significant, such that the disparity in LDKT between Black and White recipients increased with greater community-level vulnerability (ratio of adjusted Relative Risk, 0.67; 95% CI, 0.51–0.87; p = 0.003).

A similar interaction effect has also been observed in a study assessing living kidney donation by Gill and colleagues. They analyzed of data from the United Network for Organ Sharing (UNOS) registry, and found higher overall rates of living kidney donation among the African-American versus White population (incidence rate ratio 1.20, 95% confidence interval 1.17–1.24) [4]. However, this effect was mediated by income (a component of socioeconomic status), with the incidence of living kidney donation being lower among African-Americans (vs. White population) in the lowest income quintile, but better among the higher three income quintiles [4]. Their conclusion was that racial disparities associated with living kidney donation are likely related to socioeconomic factors, rather than socio-cultural factors. Our data, exploring this from a recipient perspective, corroborates these findings, and confirms a significant interaction between socioeconomic status and ethnicity for a recipient’s likelihood of receiving a LDKT.

However, reliance on data from the US may not be directly translatable to countries like the UK. Living donors in the UK do not receive financial remuneration for their donation, but are fully reimbursed for their expenses and loss of earnings during their post-operative period. Therefore, financial pressures that may dissuade potential ethnic minority living kidney donors in the US may not be translatable to other countries with universal health coverage. In the UK, Udayaraj and colleagues explored data on 12,282 kidney transplant recipients between 1997 and 2004 [5]. They observed a reduced probability of receiving a kidney from a living donor for ethnic minority recipients, and those residing in socioeconomically deprived areas, with some attenuation of the ethnic differences in their adjusted models, after controlling for socioeconomic status. However, no significant interaction was observed between ethnicity and socioeconomic status, suggesting the effect of socioeconomic status on uptake of living kidney donation is similar across all ethnic groups. This contrasts with our analysis, which found a lower likelihood of LDKT for ethnic minority recipients if they reside in areas of socioeconomic deprivation, with significant statistical interaction between the IMD and ethnicity. However, data from Wu and colleagues, analyzing data from the multi-center Access to Transplantation and Transplant Outcomes (ATTOM) study, confirms our findings of non-White ethnicity and parameters of socioeconomic deprivations to be independent factors predictive of reduced likelihood of receiving a kidney from a living kidney donor [3]. Conversely, a publication from Scotland reported no difference in the proportion of patients receiving a living-donor kidney across the quartiles of the Scottish IMD score [19].

The reasons behind the observed differences across groups of socioeconomic deprivation are speculative, and qualitative research is underway to address this discrepancy. A study conducted by Bailey and colleagues interviewed recipients who had received a deceased donor kidney, and explored the main barriers to LDKT [20]. Socioeconomically deprived recipients often reported a one-sided passive relationship with their clinician, lack of involvement in decision-making, and a lack of knowledge of the available options as some of the main reasons why they did not pursue the option of LDKT [20]. While this work suggests these patients may benefit from more targeted education about their options for kidney transplantation, it also raises concerns about socio-cultural and psychological factors that may hinder discussions in relation to living kidney donation. Ethnic minority individuals appear more likely to consider becoming living kidney donors, with higher rates of living kidney donation per million population than other ethnicities, according to national registry data in the UK [21]. This suggests transplant professionals are attaining some success in getting the message across with regards to the benefits of living kidney donation to ethnic minority communities. Our data adds to the literature by suggesting ethnic minority individuals residing in areas of socioeconomic deprivation are a high-risk group with regards to attaining a LDKT, and require more targeted focus.

Whilst we and others have reported the likelihood of receiving a living kidney donor to be lower for ethnic minority individuals or those resident in areas of socioeconomic deprivation, we do not present any data on the number of living kidney donor candidates who come forward, but fail the assessment. It is plausible that potential living kidney donors from ethnic minority communities and/or poor socioeconomic status come forward in equal numbers, but are ruled out on medical grounds. It is well known that there is a greater burden of health issues such as diabetes and/or hypertension in ethnic minority individuals and for residents within deprived areas [22‐24], which could be an important factor, and requires further investigation. Bailey and colleagues have observed potential living kidney donor candidates from ethnic minority communities have a nearly three-fold increased odds of withdrawing from the assessment process (OR: 2.98; 95% CI 1.05–8.44, p = 0.04) [25]. Although individuals from most socioeconomically deprived areas appeared to have reduced likelihood of donation, they found the trend with deprivation to be non-linear and consistent with chance (OR per IMD quintile increase: 0.88; 95% CI, 0.75–1.03; p = 0.12). In the donor and recipient sex-adjusted analysis, they found the most deprived potential donors to remain the least likely to donate, but this did not persist after adjustment for possible mediators of socioeconomic deprivation on living donation, with most IMD quintiles showing attenuation of the effect estimates. In addition, they reported that people donating to more deprived recipients were more likely to withdraw, but this association was not statistically significant at the 5% level after adjustment for donor age (OR of withdrawal per unit increase in IMD quintile: 1.13; 95% CI, 0.95–1.34; p = 0.17).

Considering our findings, and a review of the literature, we suggest educational resources to encourage living kidney donation must reflect the identified obstacles and be tailored to the individual. Strategies could include home-based discussions with allied healthcare professionals, which provide educational advice in a familiar environment, but also engages family and social support networks to aid decision-making for the potential kidney transplant candidate [2]. Importantly, such clinical trials have been shown to boost living kidney donor evaluation by over 50% [26‐28]. Streamlining the process for potential living kidney donors, and removing financial disincentives in some countries, are also important interventions to encourage more potential living kidney donors to come forward, or encourage them not to withdraw from the assessment process once started [29, 30]. The overwhelming financial benefits of successful LDKT compared to dialysis means healthcare providers have an incentive to facilitate such pathways, to ensure living kidney work-ups are streamlined and efficient for the benefit of potential candidates. Further research is clearly warranted to ensure we minimize the risk of willing potential donors withdrawing from the work-up pathway for reasons which could be amenable to intervention [31].

The limitations of this study should be appreciated for the correct interpretation of our analysis. The primary limitation was that the study only considered those patients that had received a transplant. As such, those that died on the waiting list, or that were still on the waiting list at the time of data collection will have been excluded. This may have introduced selection bias, particularly if either ethnicity or socioeconomic deprivation influence the likelihood of receiving a transplant, or the time on the waiting list. This is an important limitation which should be rectified for any future work in this area. Secondly, as a single-center study, we were not able to assess any region-specific effects; therefore the findings may not be generalizable to other centers. Thirdly, the IMD was calculated based on the recipient’s home address at the time of kidney transplant surgery. However, it is possible that patients may have changed residence during their time on the waitlist, potentially multiple times in those with long waits. Finally, the study was retrospective in nature, and so prone to all the shortcomings of such analyses, including the inability to establish causation, and the potential effects of unmeasured or intangible confounders. On the latter point, there were some variables for which data were not available, but which may have a significant influence on LDKT opportunities, including marital status and the number of siblings. There was also some granular data that were not available, such as time since dialysis commencement. The UK allocation system also evolved during this time, fundamentally after 2018 when waiting time was calculated from commencement of dialysis rather than time joining the waiting list. This confounder will impact upon the granular data for waiting time versus dialysis status.

Whilst the results of the study should be generalizable to other centers within the UK, they may not be applicable in other countries for several reasons. Firstly, the quantification of socioeconomic deprivation by the IMD quintile may not translate well to other countries, where either the baseline level of deprivation, or disparity between the most and least deprived residents, is different to that in the UK. The findings may also not be applicable to cultures with different views on the ethics or appropriateness of living-donor transplantation, particularly if these vary by ethnicity. Finally, the UK does not offer financial incentives for becoming a living donor, but does reimburse donors for their expenses. As such, the results of the analysis may not be generalizable to countries that use a different system of financial remuneration, particularly with respect to the effect of socioeconomic deprivation.

To conclude, our study identifies both socioeconomic deprivation and non-White ethnicity are being associated with a lower proportion of transplants being LDKT at a patient-level, with the effect of ethnicity being most pronounced in areas with the greatest levels of socioeconomic deprivation. This analysis highlights the importance of encouraging discussions relating to the benefits of LDKT to individuals who reside in socioeconomically deprived areas. However, it also suggests ethnic minority candidates in these deprived areas require targeted intervention to maximize opportunities to facilitate receiving a LDKT.