Liver transplant (LT) remains the definitive treatment for patients with end-stage liver disease and it is the second most common solid organ transplantation after kidney transplant worldwide [1]. Whereas the number of patients with irreversible liver injury has increased, the number of available organs has remained comparatively low creating a shortage of organs for transplantation [2]. This imbalance emphasizes the need for systems of equitable organ allocation. In 2002, the Organ Procurement and Transplantation Network (OPTN) adopted the model for end-stage liver disease (MELD) score, both for listing and recipient selection, changing the policy of organ allocation [3]. This new allocation system was aimed to ensure that patients at greatest risk for mortality are given the highest priority for transplantation [4]. Unfortunately, significant racial and ethnic disparities in waitlist outcomes have been documented during the post-MELD era and some of these disparities have been attributed to geographic differences in organ availability [5].

On June 18th, 2013, the OPTN implemented a new policy, known as “Share 35”, to reduce the number of deaths on the waitlist by increasing regional sharing of livers for patients with a MELD score ≥35 [6]. However, concerns such as the potential to increase travel distance, increased costs, and lowered liver availability in some areas of the country arose [7]. These elements could potentially negatively impact waitlist outcomes among ethnic minorities, especially Hispanics.

Therefore, we aimed to determine waitlist survival after the implementation of the Share 35 policy in non-Hispanic white and Hispanic patients. Identification of potential disparities is an important first step in developing interventions to reduce these inequalities and allows public health professionals to target populations at particular risk.

Our study has demonstrated that racial/ethnic disparities in waitlist outcomes are still present and remain an area of concern in the post-Share 35 era. These disparities extend to Hispanics who are both more likely to be removed from the waitlist and less likely to receive a transplant than their non-Hispanic white counterparts. Although previous studies have shown that Hispanic candidates appeared to fare slightly better under Share 35, their mortality was not significantly different across eras [14].

Within the constraints of a retrospective study using a large database without granular data, we provide several hypotheses for these disparities. First, Hispanic patients were listed with a higher MELD score compared to non-Hispanic whites, potentially indicating a delay in referral of these patients for transplant which may play a role in their higher rate of adverse waitlist outcomes. Such delays may be due to socioeconomic disparities like educational level or source of health care payment insurance status playing a larger role [15]. Second, we may intuitively attribute such observations to language barriers, which may impair Hispanic patients’ ability to better understand their conditions, thus making them less likely to seek medical care earlier in disease course. Third, differences in literacy and numeric skills have shown to affect individuals’ ability to obtain relevant information and their opportunity to apply that information in communications with health professionals [16]. The complexity of liver cirrhosis requires extensive self-care management that also increases the demands on patients to understand, and utilize health information [17].

Nonetheless, after accounting for these covariates, Hispanic patients remained at higher risk compared to non-Hispanic white patients by the nature of ethnicity alone. Pingitore et al. have described that a variant of the human patatin-like phospholipase domain-containing 3 (PNPLA3) gene, which encodes for a protein of 481 amino-acids, is highly expressed in the liver and strongly associated with entire spectrum of liver diseases [18]. Its mutant protein has a negative effect on triglycerides mobilization from liver droplets that leads to inflammatory infiltrate and the initial presence of fibrotic lesions [19,20]. The frequency of this variant is higher in the Hispanic population and accounts for an elevated risk for hepatic injury compared to individuals of other ethnicities [21,22]. Consequently, these inherent biologic factors could serve as an explanation of their higher waitlist mortality.

In contrast to our study, a recent study by Kaswala et al. found a lower risk of waitlist death in Hispanics compared to non-Hispanic whites (HR 0.84, 95% CI 0.79–0.90, P < 0.001) [23]. In their analysis, they used only waitlist death and not removal for clinical deterioration which generally also results in death. In addition, they used only Cox regression in their analysis and not competing risk regression which treats those who are transplanted (generally higher MELD and also higher risk of death) the same as those who are censored and does not accurately reflect the survival probability [24].

Our study also shows that a greater proportion of Hispanics are transplanted at high MELD score regions and experience longer waitlist time. However, on a competing risk analysis we found no differences in waitlist mortality between high and low MELD regions in the post-share 35 era. These results are in contrast to that of Volk et al. who found that Hispanics have not fared as well during the post-MELD era and are less likely than Caucasians to receive a transplant given that they live in transplant regions with more severe organ shortage [5]. Even though the implementation of share 35 has resulted in liver transplant being more equitable across regions, it has not addressed racial/ethnic disparities that are still present.

The strength of this study is the utilization of the competing risk analysis method, which allows the simultaneous assessment of the effects of competing risks such as waitlist removal for death or deterioration and transplant on the survival probability for each failure type. In addition, the utilization of a large database and the large scale of the study allows for a better reflection and assessment of the disparities that are currently present in the allocation of liver transplants in Hispanic populations. However, like all similar studies, this study is limited by the retrospective nature of the analysis that is limited by the available data in the UNOS database. For example, there was not available data on patient’s occupation, psychosocial characteristics or proximity to transplant centers, which have been shown in small single center studies to impact access to liver transplantation [16]. In addition, Hispanic ethnicity in our data was self‐reported and may represent a heterogeneous mix of races. Given the strong effect of race on many health outcomes, there is a risk of misclassification bias. Although the multivariate model attempts to include adjustments for potential confounders, it should be acknowledged that our study was only able to include variables that were available in SRTR and thus there remains the potential for unmeasured confounders. Finally, the SRTR database presents challenges related to accurate coding and data availability.

Our study demonstrates that despite the improvement of geographic disparities with the new Share 35 policy, racial/ethnic disparities remain an area of concern. A combination of factors such as differences in socioeconomic status, health literacy, or even genetic factors potentially play an important role. Clinicians should continue to ensure patients have access to readily available translators, health care information in their native language, and social workers with experience in ethnic minorities in order to detect and avoid inequalities in the fair allocation of scarce life-saving organs.

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

The authors declare no conflicts of interest.