Hi, my name is Dr. David Goldberg. I'm an Associate Professor of Medicine and Public Health Sciences at the University of Miami Miller School of Medicine. I'll be the moderator for this talk and the seminar, which is relating to disparities in liver transplantation. The purpose of this introduction is to give a landscape of disparities in liver transplantation, focusing on gender, pediatrics, race, ethnicity, and socioeconomic status. I have no relevant financial disclosures as it relates to this talk. The objectives of my overview are threefold. First, is to discuss potential sources of disparities in liver transplantation, to review the underlying regulatory policy guiding access to the transplant waitlist and organ allocation, and three, discuss how the current disparities may serve as violating the current regulatory framework. What is the process by which a patient gets waitlisted and ultimately transplanted? First, the patient must develop advanced and or decompensated liver disease. But this is not enough. The treating physician, usually a gastroenterologist or hepatologist, but could be a primary care physician, must recognize the severity of liver disease and the need for transplant. The patient then is referred for transplant, evaluated for transplant, and placed on the waitlist. Obviously, this is an oversimplification of what happens in reality. Once on the waitlist, there's really three broad outcomes that a patient will have. They will potentially have an available living donor and receive a living donor transplant. They will die or become too sick to transplant, then be removed from the waitlist, or they'll get high enough waitlist priority to receive a deceased donor of liver transplant. There are obviously other outcomes, the rare times a patient gets better, a patient may move, but these are really the three main outcomes of a patient on the waiting list. Now, what are the potential sources of disparities in the pathway of getting waitlisted and transplanted? A patient develops advanced liver disease, but there may be disparities in access to specialty care like hepatology care. A patient may develop new onset ascites, but may not have access to specialty care and be managed by a primary care physician, who may not recognize that this is severe liver disease and the need for transplant. I often have had patients referred to my clinic, who for several years have had complications of liver disease, and it was attributable to something else, such as severe thrombocytopenia that's followed by hematologists without recognition of advanced liver disease. Once the patient is being treated by an appropriate physician, there may be disparities in decisions about which patients to refer. There may be conscious or unconscious bias. There may be issues related to patient's socioeconomic status, social support, the etiology of liver disease, alcohol, or things like that, that may lead to disparities in referral. Once a patient is referred, they're not necessarily evaluated. Some centers have a pre-screening process, which is based on the severity of liver disease, maybe other medical factors, but also financial clearance for evaluation. There are still some states in our country that patients receive Medicaid, but Medicaid really does not pay for a transplant. Therefore, this is a source of disparity in being evaluated. Once evaluated, there may be potential disparities in being placed on the waitlist. We know that there are differences in decisions to waitlist by centers, which may be a source of disparities with respect to social support. Do patients have loved ones that can take off work, take them to appointments, and things like that? Financial clearance, both in terms of insurance, but the ability to have co-pays, especially as it relates to pharmaceutical therapy and post-transplant immunosuppression. There unfortunately remains unconscious bias in how centers make decisions. These have been studied in the setting of having people embedded in waitlist meetings, and we know that centers vary in how they make those decisions. Once someone's on the waitlist, there are then disparities in who gets transplanted and who dies. This can be related to the MELD score calculation in the formula, although recent data with the MELD 3.0 may hopefully help to mitigate this. The availability of living donors, different patients may have different access to having friends or family who can take off work and then be their living donor. Organ allocation priority, we know there's disparities in geography, but also the MELD score cutoffs, and how the MELD score formula, especially for women, may make them less likely to be in the highest priority. There are issues related to donor-recipient size matching. There's a lack of donor-recipient matching based on predicted outcomes. Unlike kidney, where patients with the best expected post-transplant survival get access to the best organs, that doesn't exist in liver transplantation. There may be unconscious or conscious bias in organ offer acceptances. I was at a talk once where someone said they don't go to fly-outs for distant organs for patients with Medicaid because if they do the transplant, it's not going to be financially net neutral to take the cost of flying out for such a patient. What is the regulatory framework of transplantation? There's really two sources. There's NODA and there's a final rule. NODA is the National Organ Transplant Act that was passed in 1984 that established the Organ Procurement and Transplantation Network to maintain the national system to match organs and individuals. Key aspects as it relates to ongoing disparities of NODA. First, NODA mentions that it's important to recognize the differences in health and in organ transplantation issues between children and adults throughout the system and adopt criteria, policies, and procedures that address the unique healthcare needs of children. Again, this will be one of the sessions in this seminar. The second, it says that the OPTN must carry out studies and demonstration projects for the purpose of improving procedures for organ donation procurement and allocation, including but not limited to projects to examine and attempt to increase transplantation among populations with special needs including children and individuals who are members of racial or ethnic minority groups and populations with limited access to transportation. Again, these are key groups that we're going to be discussing in this session. What about the other regulatory framework? We hear more about the final rule. Despite what is said a lot, there is no the final rule. It's not limited to liver transplantation. The term is actually a final rule. There are many final rules which are actually federal administrative regulations that are advanced through the rulemaking process from a rule being proposed to public comment, to being in the federal register, to being finalized. Although people talk about the final rule of liver transplantation, that's not really a thing. The Department of Health and Human Services implemented a final rule to establish the regulatory framework of the OPTN, effective March 16th in 2000. There are several sections and despite what you've heard within transplantation the last several years, it's not just about geography. There are several aspects of the final rule as it pertains to transplantation in terms of the organization, the OPTN, listing requirements, allocation, reviews and enforcement, things like that. But as it pertains to disparities, we're going to focus on a few of them. One is the comment about listing requirements. Specifically, it says transplant hospitals shall assure that individuals are placed on the waiting list as soon as they are determined to be candidates for transplantation. This is a regulatory framework, but how might disparities exist? Well, are all patient populations held to the same standard? Social support requirements may differ. Is that really a mandatory thing to be listed and are all patients held the same? What about wait-listing criteria and timeliness of wait-listing based on race or SES? Is this really being uniformly applied? The next is allocation of organs. Again, Section 121.8 is actually not just about geography. What takes precedence in policy development? Aspects of sound medical judgment, best use of donated organs, avoiding wasting organs and promoting patient access to transplantation. Again, access as a source of disparities. As it gets to the allocation of organs, there are a couple of things and I apologize for the slide. With respect to policy development, we must promote access to transplantation. We would ask as we go through these seminars, is access to the wait-list and transplant the same for all patients? For example, race. Allocation performance goals are another thing to think about. The final rule says, standardized criteria for determining suitable candidate through use of objective and measurable medical criteria for adding or removing patients. Something to think about is our criteria for wait-listing or removing patients, i.e. too sick to transplant, the same for all patients. I list women because there's been studies showing a disparity here. Also, we must set priority through objective and measurable medical criteria, and ranking shall be ordered from the most to least medically urgent. As we hear the next set of talks to think about, does male-based allocation accurately rank order patients? I'd like to just briefly introduce you to our tour guides that are reviewing the landscape of disparities. First off, we're going to be hearing about gender equity with a focus of renal function and donor-recipient size mismatch. This talk will be given by Elizabeth Betsy Verna, who's an Associate Professor of Medicine at Columbia University. She's a researcher in gender disparities and was a co-author of a JAMA surgery editorial entitled, Time for Action to Address the Persistent Sex Disparity in Liver Transplant Access. Next, we'll hear about maximizing pediatric allocation. Dr. Evelyn Hsu is an Associate Professor of Pediatrics at the University of Washington. She's a researcher and a fierce patient advocate for pediatric liver patients and for recognizing that the needs of pediatric patients are different than adults. Then we'll be hearing about racial disparities in post-transplant outcomes from Dr. Kimberly Ford, Associate Professor of Medicine at Temple University. She's a disparities researcher who for a long time has been studying racial and gender disparities across the spectrum of transplant. Lastly, we will hear about transplant care to the uninsured and underinsured from Dr. Nora Turow, Professor of Medicine at the University of Southern California, who recently published a seminal paper on the issues of transplantation for unauthorized immigrants in the journal Hepatology. With that, we'll go on to the remainder of the talks. Good afternoon. I'd like to thank the Liver Transplantation and Surgery SIG for inviting me to present on gender equity and transplantation, including a focus on renal function and donor-recipient size mismatch. My name is Elizabeth Verna and I'm from Columbia University in New York. MELD-based liver allocation was adapted in 2002 to be a more fair and objective system to prioritize patients on the waitlist. However, even in early analyses of the impact of MELDs on waitlist disparities, it was acknowledged that the system might have accidentally exacerbated sex-based disparities in transplant outcomes. Here in this analysis comparing four years of waitlist registrants prior to the implementation of MELD to four years post, women had a new 30 percent increase in the odds of death or becoming too sick on the waitlist, and significantly lower odds of transplant within three years. This disparity in waitlist outcomes and increased mortality has been demonstrated multiple times since then by several different groups. Here's an example in data that goes up to 2008, where holding MELD constant, women still had a 19 percent higher risk of waitlist mortality. When plotted out over time, it is clear that although there have been tweaks in the MELD score to address some of its insufficiencies for some groups of patients, that there's really been no change in sex-based disparities in terms of access to transplantation or rates of transplant comparing women to men. In this analysis, if you assume that women and men should have the same access to deceased donor organs and therefore equal rates of waitlist mortality, there may be over 800 women's deaths over the last decade that could have been prevented if we had addressed this disparity sooner. Since the initial recognition of this disparity, a key feature that's been discussed is the selection of creatinine as the measure of kidney function in the MELD score. This perhaps should have been controversial anyway as serum creatinine clearly underestimates kidney dysfunction in patients with decompensated cirrhosis. But it's also clear that the relationship between measured GFR and serum creatinine depends on many other variables as well, including age, race, and also sex. Therefore, creatinine values between men and women are really not directly comparable. This has been shown in many different circumstances, but this is a figure from some of the original MDRD publications, where they showed that for a GFR of 60, this correlated with a serum creatinine level of 1.3-1.5 among white men, but 1.0-1.2 among white women. In fact, the differences between men and women widen as kidney dysfunction progresses. Therefore, and this has been shown by many different groups, women accrue fewer points, perhaps one to three fewer MELD points from serum creatinine and starting at a later stage of kidney dysfunction than men with the same EGFR. It's pretty clear that this is likely an important source of this sex-based disparity in waitlist outcomes. However, it's not the only source as that even for the same MELD score, women are still significantly less likely to be transplanted. What else is going on here? I would say another major theme of the literature is that body size and perhaps particularly height is also probably having a significant impact on transplant access. In this paper, when they were looking at the multiple different factors that could be contributing to sex-based disparities and waitlist outcomes, they noted that women were more likely to be of short stature, below 165 centimeters, where men were more likely to be above 165 or even 180 centimeters on the waitlist. When they adjusted for MELD score, they found that there was a 19 percent increase in the risk of waitlist mortality for women. When they further adjusted for other things known to contribute to waitlist mortality, like blood type, age, race, and cirrhosis etiology, that there was still a significant increase in waitlist mortality risk. However, when they further adjusted for height, that actually much of this risk was mitigated, arguing that perhaps height was an important mediator of the relationship between sex and waitlist mortality. This group then went on to try to determine the optimal cutoff below which people were significantly disadvantaged. They found that a height of 166 centimeters or about five feet, five inches was the height below which people are at significantly higher risk of either dying or being delisted on a transplant list and also had lower rates of transplantation. In fact, after adjustment, this height of less than 166 centimeters remained associated with an eight percent increased risk in waitlist mortality. Not surprisingly, women were more likely to be below this height cutoff. In fact, 72 percent of women actually were below this height cutoff, placing them at increased risk of waitlist mortality. Why is it that height would contribute so significant to waitlist outcomes? Perhaps the most direct reason that might be is that size mismatch is a very common reason to have organ offers declined. As we all know, a substantial proportion of livers are not actually allocated to the candidates in the top three positions on the match run and one of the most important reasons for this is actually donor recipient size mismatch. This group actually showed that women are significantly more likely to have an organ offer declined in the first, second, or third positions on the match run. When they looked at why that might be, women were actually also more likely to be declined specifically for the reason of donor recipient size mismatch when that's represented with these multiple different ways that can be coded. Still, organ offer declines were more frequent among women of smaller stature than among men of smaller stature. It's not just the stature at play here, and I think that's an important thing for us to remember. However, when they did a mediation analysis, they also found that height and other body anthropometric measurements were important mediators in the relationship between sex and the increased risk in organ offer declines. Finally, I just want to note that there might be other factors at play here too. Women are not only more likely to die in the wait list and less likely to be transplanted, but they're also more likely to be delisted. Why might that be? This is even when you adjust for age, height, melt sodium, performance status, and region. Is it because women are more likely to be perceived as frail? Is there some implicit bias at play here? Is there some reason that the natural history of cirrhosis on the wait list would be different between women and men? I think there's many other factors that need to be investigated further. Finally, in the most recent update of UNOS analysis from 2013 to 2018, I think it is really clear now that there's many different factors at play that accumulate to form this increased risk. In this paper, when they adjusted for geographic region, melt score, anthropometric, and liver measurements, they found that altogether, these women have a 25 percent greater risk of wait list mortality compared to men, which I think is really just an unacceptable disparity in wait list outcomes that has not yet been addressed. So now that we've demonstrated these disparities over the last decade, what are we going to do to fix them? And there's many different things that have been suggested over time. And I put them here into sort of three different buckets. The first thing is to fix the MELD score. And there's many different possible ways to do that, including perhaps substituting creatinine for a calculated GFR, adding additional points for female gender or for height, or perhaps going back to the drawing board and creating a new MELD score like MELD 3.0, which we'll talk about in a few slides to come. Other things that have been suggested include figuring out a way to preferentially get smaller graphs to either women or people of short stature, perhaps smaller deceased donors, pediatric donors, or increased use of partial graphs like split livers and living donor transplant. And finally, additional actions, which again, I don't have time to cover here in detail, include things like studying the impact of implicit bias, increasing the number of women professionals in the transplant field. And I think very importantly, studying the impact of gender upon metrics upstream of transplant listing. So what about substituting EGFR? And there's many different EGFR calculators out there, including liver specific EGFR calculators, such as the GRAIL score. And when GRAIL is combined with other factors in the MELD score to create here something like the MELD GRAIL sodium, it's pretty clear that these measures more accurately predict weightless outcomes, especially among high MELD patients, and among women than the creatinine based system. However, I think implementing changing creatinine to EGFR is probably going to be quite complicated. In a recent systematic review and meta analysis of over 25 studies with 18 equations, I think we were again reminded that many of these different EGFR estimators still don't perform optimally in populations that are particular interests on the transplant weightless, including those with the EGFR under 60, and those with ascites. And in fact, cystatin C based estimators are probably going to be the most accurate as they're less impacted by cirrhotic physiology. However, that would be very difficult to implement as that's not widely being measured these days, and certainly would be difficult to model retrospectively, given we don't have that in the UNOS datasets. I also just want to point out here that most of these high-performing EGFR equations also include variables for race. And of course, we would want to make sure that we didn't inadvertently create an additional disparity for other populations while trying to fix this one related to sex. So what about going back to the drawing board and reconceiving a new MELD score based on more updated data, looking at 90-day survival in a more recent cohort. So Ray Kim and the group out of Stanford recently published this where they looked at UNOS data from 2016 to 2018, and looked at all the different subjective variables in the UNOS dataset to predict 90-day survival. In their final model, they included the additional variables of sex, albumin and interaction terms between bilirubin and sodium and creatinine and albumin in order to create a new MELD 3.0 calculator. They made some additional tweaks to the model including capping creatinine at 3.0, and also made the case that substitution of GFR should not be considered in some ways for the reasons that I just articulated. They also made the case that they wanted to include sex as one of the variables but not height, and that sex and height were actually very collinear. And that in fact, as you can see here in the figure that among men, height actually didn't impact weightless outcomes, almost at all, where among women, it very strongly impacted outcomes. And then overall, sex had a larger impact on the model. So that was the variable that they kept. MELD 3.0 better discriminated 90-day weightless mortality compared to MELD sodium in their validation cohort. They also made the case that they could have correctly reclassified a net of almost 9% of the people that died into higher MELD tiers, perhaps giving them better access to transplant. Interestingly, it was almost 15% of the women who died versus only 4% of the men. And then an LSAM analysis showed that MELD 3.0 resulted in fewer weightless deaths compared to MELD sodium. However, whether MELD 3.0 is ready for primetime or not, I think could be debated, and it's also outside the scope of this talk. But I think these types of changes, really pragmatic solutions that we could implement relatively quickly into the system are things that we really need to strongly consider at this time. What additional policy considerations could we think about to, for example, preferentially allocate smaller organs to women or those of shorter stature? So here's an example where, again, the group out of UCSF showed that women who receive a pediatric liver donor as their first offer have a lower risk of weightless mortality compared to those who receive an adult offer first. And again, this might get at the issue of turndowns and size mismatch. And they then went on to do LSAM modeling of a potentially new allocation system, where pediatric donor livers were preferentially offered to adults less than 166 centimeters within each MELD score band and within 500 nautical miles, and importantly, after all pediatric candidates were exhausted. And through this modeling, they found that it's possible to perhaps eliminate a substantial portion of the sex-based gap in the proportion of adults who receive liver transplantation. And finally, another way to increase utilization of small grafts for this population is to think about increases in both split liver transplantation as well as increased use of living donor liver transplant. And in this very interesting study out of Toronto, they showed that women who had a potential living donor actually had better weightless outcomes than women that did not have a potential living donor. And when they compared them to men, men and women with a potential living donor actually had similar weightless outcomes, where women that did not have a potential living donor had worse outcomes compared to men without a potential living donor. And I think this again raises the issue that we need to really think about increasing the use of smaller grafts, particularly among women and particularly among women of shorter stature. So in summary, women on the transplant weightless face persistent and I think really unacceptable disparities in access to transplant. They're significantly less likely to be transplanted and maybe at as much as a 25% increased risk and dying. There are many reasons for this, including the fact that we use creatinine in the MELD score, and also their body size, but probably many other reasons as well. And really, I think urgent action is needed to implement pragmatic policy solutions to make changes to address this disparity. And I'll just conclude by saying that MELD-based allocation was established to form an objective and fair system that prioritizes illness without bias. However, for any two subjects with the same MELD or MELD-sodium score currently on the list, in fact, their access to transplant is not equal. And sex is a major driver of this. Ongoing study to address these disparities is really critical. Through refinement in our allocation, prioritization and distribution, we need to work to address these disparities. And disparities in our system are likely to persist without attempts to address larger societal issues as well. Thank you for your attention, and I look forward to the rest of the session. So I want to thank the organizers of ASLD and of this session for inviting me to speak. I understand that this session specifically addresses disparities in access to and outcomes of liver transplantation. I'll be speaking about children on the waitlist as a whole, as a subject of the session, but I really think it's important to acknowledge that although I'm not going to focus on these issues specifically, racial and socioeconomic disparities are present and often amplified within the pediatric cohort. The title of this talk is Let No Child Die on the Waitlist, How to Maximize Liver Allocation to Pediatric Recipients. I've been doing this for some years now, and I have to say that although our community has made some progress, we haven't really made significant gains in eliminating preventable death on the liver waitlist for children. My plan today is to outline some of the obstacles that remain and hopefully to introduce solutions that will clearly allow us to achieve our goal while improving outcomes for everyone who is awaiting liver transplant. But first, disclosures. I have no relevant financial disclosures to this talk. I have received research and grant support for studies that enroll patients with pediatric liver disease that are listed here. And it's also of note that I hold leadership positions within UNOS and SPLIT, representing organizations with a stated mission of advocating for children. My hope is that you leave today's talk understanding the current challenges facing children on the liver transplant waitlist, that you understand the ethical mandates, governing allocation policy, and why they exist, and that you'll know strategies that have the greatest potential to maximize liver allocation to pediatric recipients without significantly affecting adult waitlist outcomes. So the problem is that there are not enough livers to transplant every person who needs one. And so therefore, people will die waiting for a life-saving liver transplant. The best solution to this problem solves the problem of pediatric waitlist mortality with minimal impact to adults. Ultimately, we are all on the same side. We're on the side of humans. No one human is worth more than any other human. Every human has infinite worth and is equally worthy of liver transplant. We are as a society bound by an ethical code to protect those who are most vulnerable. We are accountable to society for upholding this code. And we can fulfill our obligation to this code by finding a solution that brings benefit to all those that are waiting. I believe, and the data supports this, that children waiting for liver transplant, particularly the smallest and the youngest children, are much more vulnerable to death on the waitlist. A typical pediatric patient awaiting liver transplant has end-stage liver disease from biliary atresia, a disease of infancy. These patients comprise about 50% of children who are waiting. They have progressive and incurable liver disease, and without transplant, they decline and die. We see it again and again. This child's picture was shared by her family in hopes that they could help other children avoid the same fate. She's young, she's small, she's incredibly sick with decompensated chronic liver disease. She's had bleeding. Options to help her, like tips, if you're lucky enough to be at a center with pediatric specialists able to perform them in children this small, are challenging, complicated, and prone to clotting. 10 kilograms is the weight of an average weight of a 15-month-old baby. We can't do banding in children under this weight because the banding apparatus cannot be passed down their pharynx without causing significant injury. Because children this age maintain their cardiac output through heart rate, we're not able to use beta blockade. And finally, grafts are size-limited, so we wait. We have grown accustomed to supporting their clinical state with the placement of central line and starting parenteral nutrition, trading off the inherent risk of dying from infection with continuing to support their growth and brain development. We administer diuretics to manage fluid overload and anasarca that can have long-lasting effects on their hearing. In order to save their lives from infection, we repeatedly give them broad spectrum antibiotics, increasing their risk for colonization with resistant organisms. We draw blood and we cause iatrogenic anemia on top of their chronic anemia from blood loss. We repeatedly administer anesthetics for multiple procedures that they require, which has unknown effects on their long-term neurocognitive development. And we subject them to radiation when we screen for life-threatening infections. And then near the end, we put them in the intensive care unit. We sedate and we intubate them. We place them on dialysis and plasma exchange to maintain a tenuous window of stability to allow for transplantation. And we wait. If they're lucky enough to be transplanted and we transplant them in time, their disease never recurs and they have the best outcomes post transplant with over 90% 10 and 15 year survival. So here's the data. The smallest children have the highest overall death rate. This is represented by the bar on the far left. So you can see that the death rate per 100 patient years is above that of any other age group. It's 25% higher than the death rate of all adults on the wait list. And this is most of the children that we take care of. Children under five years of age comprise 62% of pediatric liver transplants. The pediatric mortality rate as represented by this lavender line improved significantly about 10 years ago, but has really plateaued in the last decade, remaining nearly twice that of all other age groups combined. This plateau has persisted over time, despite increasing use of splits, high rates of exception scores and living donation. And what we've also found out is that these kids are getting sicker. The acuity and degree of illness a transplant has continued to rise over time implying that these children are more ill before we're able to transplant them, and we're waiting longer. We know that the pediatric end-stage liver disease score, the score that puts kids in order for transplant, as compared to the MELD for adults, underestimates children's mortality on the liver wait list. So the bottom gray line in the graph on the right side represents how the PELD score predicts expected mortality. The dark blue and the orange line represents how actual mortality in the full and reduced cohorts of the SRTR database is. You can see here that the actual probability of death, particularly at calculated scores above 10, and much worse at scores above 20, is much higher than that, that the PELD score was designed to predict. This is why we've had to use exception scores. These children really don't stand a chance. This graph from the most recent SRTR liver report shows that pediatric donors have been steadily decreasing over time. You can see in the lavender line, a steady decline that has persisted from 2006 on. The green line of 18 to 34 year old donors prime graphs for splitting has remained high, and there's been a steady increase as well in the 35 to 49 year old donors, as represented by the pink line, who also arguably could be used for splitting. And we know the children still show up on the same match runs as adults for organs. So what is the solution? The solution is really uncomplicated. There's a limited menu that we have to pick from. And the punchline is that in order to keep kids from dying preventable deaths on the waitlist, we have to give them timely access to any suitable graph. And we have to be willing and skilled enough to transplant them with those graphs. Both of those things have to happen. It isn't or, it's and. As I previously stated, this is not about declaring one group of humans more worthy than any other group of humans. Every human has infinite worth and is equally worthy of liver transplant. But ultimately, we are definitely bound to care for those patients in front of us to the best of our ability. But we're also lucky enough to exist in a system that has a moral and an ethical code. This code has delivered a directive to us on creating allocation policy and the directive with regards to children is clear. How are we asked to make decisions by these governing bodies? To answer, we must refer to the legal documents that established organ allocation in this country. In 1984, the Organ Procurement and Transplantation Network was tasked by the National Organ Transplant Act to recognizes the differences in health and in organ transplantation issues between children and adults and to adopt criteria, policies and procedures that address the unique healthcare needs of children. In the year 2000, the HHSS implemented the final rule, which declared that allocation be equitable and achieve the best use of donor organs. Protection of children really within society really became into being in the last 100 years, the Declaration of the Rights of the Child was endorsed by the League of Nations in 1924 and adopted by the UN General Assembly in 1959. Within this declaration, they recognize that providing resources to protect children and help them thrive is a fundamental human instinct. And that the children need special safeguards and care, including legal protection because of his physical and mental immaturity. Laws that enable children to develop in a healthy, normal manner. And in enacting laws, the best interests of the child shall be of paramount consideration. The UNOS Ethics and Pediatric Transplantation Committees took this a step further in 2014, when they published the ethical principles of pediatric organ allocation, outlining four ethical arguments in favor of pediatric protection. They acknowledged that early transplantation to children addresses time limited opportunities to benefit from transplant. They also recognize that we ought to maximize opportunities to allow a human to reach a full lifespan and children don't have as much of an opportunity to do so. They also recognize that children are the least advantaged group and that we should maximize even the minimum benefit that they could have. And only lastly, did they consider that overall patient survival is markedly greater for pediatric liver transplant recipients. It's possible that if we transplant all of the children, the number won't go up. There isn't this other mass of children with liver disease that require liver transplant. If you look at the lavender line that's really hidden at the bottom of that graph, that represents all of the patients under 18. And remember, the majority of these recipients are small babies. The number comprises a very small amount of total transplants, and it has been steady for two decades and we don't anticipate an increase. This is unlike the adults on the list for whom transplant numbers have been increasing, which is of course a good thing to allow this life-saving therapy to be accessed by all of our citizens. It's arguable to say that if we transplant all, we'll transplant them all. And this has transpired in all other countries that have implemented solutions to do so. In fact, as we inch ever closer to finding answers and treatments to diseases that were previously a mystery, such as treatment for immune-mediated acute liver failure, gene therapy for rare diseases, novel treatments for cholestatic liver diseases, we may see a decrease in the total number of children requiring liver transplant, which brings us back to solutions. Looking outside the United States, we have seen that countries such as France, the UK, and Italy have implemented a combination of pediatric priority and mandatory splitting to effectively eliminate their preventable pediatric waitlist mortality while increasing total numbers of transplants and maintaining excellent outcomes for everyone. So the solution could be that we prioritize children. The current PELD score addresses discrimination, ranking children against each other, but not directly calibration, ranking children with PELDs against adults with MELDs. We know it underestimates mortality. We need it to accurately estimate mortality, but we have the ability to set the score and adjust it to what we believe the mortality rate of children should be. This graph here represents US data on mortality rates for 2018 and 2019. Children don't die at the same rates as adults in the general population, so should they die at the same rate as adults on the liver waitlist? This graph, generated by researchers at the SRTR, represents how the PELD score in yellow dots predicts mortality for children at any given score. The MELD sodium age standardized score is represented in orange, how the score performs to predict mortality in adults on the list. You can see here that for any given score, the original PELD performs poorly, particularly for scores 10 to 40, under predicting mortality in comparison to adults. The green dots represent the PELD creatinine, the likely new proposed score. This score has three points added to all PELD scores to adjust for this poor comparative performance, but why stop at three? If we really wanted to incorporate priority, we could just add 20 or 30 more points. This could also be achieved with a new continuous distribution allocation system. We've already discussed where we could do it by a medical urgency, but we could also add points in the patient access category, allowing for a higher composite allocation score. The second solution is splitting, or as I like to refer to it, the win-win solution to the zero-sum game. We know that long-term graph survival rates in both adults and children are comparable with those from whole organs. Pediatric split recipients, in fact, have a lower wait time and improved survival. Over five years, when more than 2,000 livers met the noted criteria for splitting, under 4% were utilized for splits. 96% were used as single whole organs in adult recipients, and a really small number of centers were doing almost half of all the splits. I think it's time to mandate split liver transplants. It's a simple solution. This solution allows us to also transplant smaller women who have disproportionately increased mortality. I've summarized the relevant solutions. We prioritize all children through age-adjusted mortality to the PELD score or continuous distribution. We mandate splitting while keeping up skills and access. Pediatric centers have to be able to do living donor liver transplant, and we have to train and sustain the future generation of surgeons to assure that this access exists. In conclusion, I know we can do this. Children have the highest vulnerability to death on the wait list. As a society, we are bound to ethical codes to protect children, and solutions are available and can be executed with minimal impact and likely benefit adults on the wait list. I do want to thank all my colleagues in the community and within UNOS who have done this work and generated the data required, to all those who continue to advocate for children, and to those who allow us the venue and platform to talk about it, and for everyone who is fighting for their patients to benefit from the miracle of transplantation. I want to thank the patients and families in the community who have been raising their voices and sharing their stories. They are the reason that we are all in this. Thank you so much. I'd like to take this opportunity to thank the organizers for the kind invitation to speak today. My name is Kimberly Ford, and I will be discussing ameliorating racial disparities in post-liver transplant outcomes as part of the session on evaluating disparities in access to and outcomes of liver transplantation. I am an Associate Professor of Medicine in the Department of Medicine, Section of Hepatology and Liver Transplantation at the Temple University Hospital at the Lewis Katz School of Medicine. I'm a board certified gastroenterologist and transplant hepatologist, and furthermore, clinical epidemiologist with expertise in utilization of large databases and administrative databases, and I'm a clinical epidemiologist with expertise in administrative databases and have clinical interests as well as research interests in viral hepatitis and disparities. I have no financial relationships to disclose. The outline for my session today includes a presentation of a simplified conceptual framework for racial disparities in post-transplant settings. I will then discuss the landscape of post-transplant disparities, including recent data on patient survival and graft survival. Further, I'll be discussing mechanisms of post-transplant disparities, including what is known about patient level and structural factors. Lastly, I will review some opportunities for improvement and for research. As this audience is well aware, liver transplantation is the only curative therapy for end-stage liver disease. The ethical framework of healthcare and more specifically transplantation focuses on equity, and we see this principle guiding paradigms in our transplant policy, including the adoption of milk-based allocation in February of 2002, and subsequently the adoption of acuity circles in 2020. While changes in allocation have certainly improved some aspects of disparity, we know that racial and ethnic disparities remain. Race and ethnicity are social constructs without a biological basis. However, disparities are a result of cumulative effects of structural processes and other socioeconomic barriers. There are clearly opportunities for remediation, which when addressed, will improve outcomes across the board. This is a simplified and perhaps oversimplified conceptual framework presented by Mather and colleagues that reviews some of the salient areas in which racial disparities play out in the process of not only evaluation for liver transplantation, but even prior to that stage when patients are presenting for care of their chronic liver disease. Today, I will be focusing on the steps between liver transplantation, and short- and long-term outcomes post-transplant. Many potential reasons have been proposed for continued existence of racial disparities, including limitations in patient resources, social support system, access to immunosuppression, medical comorbidities, and quality of specialty care in the post-transplantation setting. So, while most people would think that once one surmounts limitations in access to liver transplant, that outcomes are equivalent for patients of various racial and ethnic groups, that's actually not the case. We see that patient survival and graft survival are worse in racial and ethnic minorities. In a study published by Quillen and colleagues in 2015, 12,445 patients undergoing liver transplant from 2007 through 2011 were assessed. The effect of race on graft outcomes and patient outcomes was explored using mixed-effects proportional hazards modeling. Important factors were controlled for, including recipient and donor characteristics, geographic region, donor service area, and individual center. These researchers found that death and graft failure were increased in Blacks. There was a 31% increase in death and a 28% increase in risk of graft failure. Additionally, this analysis calculated one, three, and five-year estimates of survival. And so, if we look over time, you can see that disparities are not only emerging, but widening eight- and five-year transplant outcomes are examined. When you adjust for recipient factors, donor factors, region, donor service area, we see that really there's an attenuation of the disparity that we see here for African-Americans. But the disparity, of course, is not completely ameliorated by adjustment for these factors. There are certainly various lines of evidence which have been proposed as a rationale for disparities in patient survival or differences in survival outcomes. These have included things such as disease etiology, so a higher burden of viral hepatitis, especially hepatitis C, which was previously the primary indication for liver transplantation. And in the context of hepatitis C, recurrence was difficult to treat with the prior standard of care, particularly for African-American patients. However, with the widespread adoption of direct acting antiviral agents, this was thought to be a rather addressable issue. We also see a higher burden of HCC in minority populations. So, it's been suggested that some of the disparities and outcomes in the post-transplant setting are due to that factor. So, how are we doing? Well, this is recent SRTR data from 2019, and this report highlights the fact that overall we're doing quite well. So, 8,896 liver transplants were performed in that year. That represented more transplants than any single preceding year. This was also about an 8% increase over the past year. For racial distribution for those transplanted in year 2019, we see that approximately 71% of patients were white, 16% Hispanic, and 7.3% African-American or Black. However, in the best case scenario, we see that with outcomes in living donor transplant, over the study period, we see that racial minorities are not doing well. So, worse patient survival for Black or African-American patients. And we see the same trend here with respect to graft survival. So, what are the potential mechanisms of post-transplant disparities, both for patient and graft survival? The first of these is really disease etiology. And so, this is data from Molinari and colleagues. They examined adults requiring a cadaveric liver transplant for HCC between the years of 2002 and 2013. In this study, they performed a survival analysis that was stratified by recipient ethnicity using the Kaplan-Meier method. They then used proportional hazards modeling to assess the effect of predictors on survival. So, if we look at the, so, if we look at the post-transplant survival out to one year, we actually do not see any differences in mortality here. However, if we look out to five years, we see that African-American patients had an increased mortality. So, that's about a 50% increased mortality after adjustment. So, the take-home points here were that there were no differences in short-term outcomes based on ethnicity. African-Americans had the lowest five-year survival in the cohort, and there was a 52% increase in mortality found in adjusted analysis. In a recent study conducted by Lee and colleagues in liver transplant recipients between 2002 and 2018, logistic regression analysis and Cox proportional hazard models were used to calculate differences in post-liver transplant mortality. Interactions between transplant year and race were also examined. In addition, there was a mediation analysis that I thought did a really good job of identifying some of the differences based on disease etiology. And you can see that over the study period, there were certainly differences in mortality, but that the most profound difference in mortality was really seen in the era from 2014 through 2018. And in this mediation analysis that I just alluded to, listing for a primary diagnosis of alcohol was actually associated with the most profound effect on mortality, and that effect was most profound from 2010 through 2018. So in summary here, we see Blacks having a higher risk of dying over the... Differences in mortality were actually worsened over the period of 2014 through 2018, and alcohol-associated liver disease was the strongest mediator of that disparity over the study period. Other patient-level factors include donor-recipient race matching, which is really a controversial issue. So this is a relatively new study published by Silva and colleagues examining adult African-American patients with HCC undergoing liver transplantation from the years 1994 through 2015. Patients were separated into cohorts based on donor-recipient race matching. These matched and unmatched cohorts were thereafter followed to determine the overall survival as determined by the Kaplan-Meier method, and of course multivariate regression modeling was also performed. So these researchers found that those patients who were matched on race actually had an improvement in overall survival. So matched patients had an improved five-year survival, so that was 64% versus 56.9%, so that was a statistically significant difference. And race-matched transplantation was independently improved or independently associated with improved survival over the study period. There are other patient-level factors as well, which I'll briefly address given time constraints. These include an increase in the risk of rejection, so a higher incidence of acute rejection has been observed in Black patients. There's an increased risk of chronic reduction as well, and some research has postulated that there are polymorphisms in the cytochrome P450 enzymes which metabolize immunosuppressive drugs. Additionally, infection may be playing a role, so certainly in patients who were transplanted for hepatitis C-associated liver disease, there was an increased risk of CMV infections. I just wanted to turn briefly to some of the structural factors that may be associated with disparities. So this is a study by Ross Driscoll and colleagues that was published in liver transplantation. These researchers examined variation between center and within-center disparities among liver transplant recipients from 2010 through 2017. Proportional hazards models were used to calculate transplant-specific hazard ratios and hierarchical survival analysis was also conducted to examine whether there was an interaction between race and survival. Within the context of this, there were various center-level characteristics that were controlled for. These investigators found that Black patients experienced 1.11 excess deaths after transplant. They found that variations in disparity were seen across different transplant centers. They also found that this variability was not explained by center-level characteristics that we tend to focus on in our research, including volume, proportion of minority patients served, quality rating, or region. This is a really provocative graphic where we see low disparity centers, so really no difference in post-transplant outcomes between minority patients and white patients, and then high disparity centers. Clearly, there's something operative here and further research is going to be needed to identify what these specific aspects of the transplant center are in order to address racial disparities. The takeaway points from today's session include racial disparities in post-liver transplant outcomes have remained consistent over time and even increasing in the recent past. Disparities exist in patient-level factors and structural factors. More research, of course, is required to ameliorate racial disparities in post-liver transplant outcomes. Just some thoughts about potential research. They include certainly quantitative research, such as examination of post-liver transplant outcomes by disease process and we see some of the work being done here with alcohol-associated liver disease, but certainly more could be done with respect to hepatocellular carcinoma and examination of post-liver transplant outcomes by allocation system. How does the implementation of acuity circles affect racial disparities post-transplant? Then some qualitative research is also needed. A comparison of those high and low disparity transplant centers and really an exploration of quality of follow-up care. These are all going to be needed going forward to address racial disparities in the post-liver transplant center. Thank you. I'd like to thank the organizers for the kind invitation to participate. I'll be addressing transplant care in the uninsured and underinsured. These are my disclosures. Looking at the insurance status among transplant candidates and recipients is insightful. First of all, looking on the left, you can see that of transplant recipients, about half have private insurance and half have government-supported programs, Medicare, Medicaid, and the VA. Then looking on the right, you can see that the proportion with private insurance versus Medi-Cal and Medicaid is changing over time, with about a 1-1.5 percent decline in those with private insurance per year since 2010-2018, and approximately similar increase in the proportion that are utilizing Medi-Cal or Medicare. Over time, private insurance on the decline. There's substantial regional variation in terms of where those individuals with Medicaid insurance reside, with Region 5 having the highest proportion. One out of every four patients who've undergone liver transplant in the US with Medicaid coverage are in Region 5. But if you look by region in terms of the proportion of their patients that are covered by Medicaid, it's more uniform as you can see here. Region 4 has the lowest at 9 percent, but most regions have approximately 15-20 percent of their patients that have Medicaid insurance and would be viewed as underinsured. So why is it important? Well, we recognize that there are very many studies that show that there are higher rates of waitlist removal and mortality in patients of public insurance. And shown here are data by payer type, and you can see that those with private insurance are more likely to be transplanted, less likely to be removed due to being too sick or for other reasons. Indeed, in a multivariate model of waitlist outcomes, Medicare had a 19 percent lower odds of transplantation, Medicaid 24 percent lower chance of receiving a liver transplant. Similarly, post-transplant survival is influenced by insurance status. Shown here are post-transplant five-year mortality, and although overall mortality is reduced across all candidate types in terms of the insurance coverage, still there's a disparity in that there are higher rates of mortality in those that have Medicare or Medicaid insurance. And that's, again, shown in the table on the right where the adjusted hazard in a fully adjusted model shows that Medicare patients have a 24 percent higher rate of post-transplant death, Medicaid 14 percent higher risk of death compared to those with private insurance. And we also know that if you transition from private to public insurance post-transplant, that that also is associated with worse outcomes. Shown here is graft survival in contrast to patient survival, but looking at individuals who had continuous private insurance beyond the first year post-transplant versus those that either had continuous Medicaid or Medicare insurance, and importantly, those that transitioned. And you can see that anyone who either transitioned or had continuous Medicaid or Medicare insurance in this analysis had a worse outcome compared to patients who had continuous private insurance. So why is it? Well, that's where I think there's still much more research that needs to be done. While it's been shown many times that there's worse outcomes, understanding the why is still, I think, in its infancy. And shown on this slide are the various purported factors that might be important. Highlighted in red is the importance of access to care, both access to primary care for comorbidity management, as well as access to specialty care and the coordination between those specialists and primary care providers. There's also issues related to just social program access or other medical service access that's important. And then increasingly, we recognize that the social determinants of health are probably relevant as they substantially overlap with patients who access public insurance. So understanding these is key to how we deal with the disparity that exists between those with private and public insurance that improve outcome. An important aspect of providing transplant as an option to the uninsured is through Medicaid expansion. In 2013, approximately 44 million Americans were uninsured. And in 2014, the Affordable Care Act sought to expand insurance coverage through an optional state level expansion of Medicaid. And here in 2021, you can see that 39 states have now expanded to include this state option. This important study examined Medicaid expansion and change in waitlist registrants. It's published in JAMA, Open Network 2020. Shown on the left are the estimated number of waitlists listed per 100 person years between 2010 and 2017, with 2014, the time of expansion, shown in the blue line. And you can see that the expansion states in black and non-expansion states in orange both saw an increase after 2014. But really, an important caveat is that around 2014 is also when DAAs became available. And so that was strongly influencing the HCC referrals for liver transplantation. So if you look at this with HCB excluded, as shown in the inset, you perhaps get a slightly different picture, but again, see that in both expansion and non-expansion states that there was a increase in the number of waitlist registrants. But what really matters here is does that annual percent change differ by insurance type? And that's shown in the table on the right, where you can see for the privately insured patients that there was an increase in both extension and non-expansion states with or excluding HCV. But the really striking difference is when you look at the Medicaid group. And here, you only see an annual percent change in the incidence rate ratio among those that were in the expansion states and not seeing in the patients that were in the non-expansion states. This study also looked at the effect of Medicaid expansion among the Medicaid patients by race ethnicity and showed that it was not equal across all races. So shown in the figures, which look at expansion, actual versus predicted in red versus non-expansion in blue, you can see that the Medicaid expansion was associated with increased rate of listing for liver transplantation in whites and in Hispanics, but not seen for blacks. So understanding this differential or the benefits of Medicaid expansion is still an important area for future research. Turning now to the post-transplant care and whether Medicaid expansion has made a difference. This is a study looking from 2009 to 2013 and looking at participation in Medicaid post-transplant. Shown on the left is a multivariate competing risks analysis of individuals with private insurance at transplant and then their likelihood of having Medicaid participation post-transplant. And then looking at the effect of being in a Medicaid expansion state versus not. So shown is that indeed there was an era effect with a 50% higher likelihood of having first participation in Medicaid in this private insurance group. And there was definitely an interaction between those that were in states with and without Medicaid expansion. You can see in the non-expansion state that the hazard ratio includes one and does not show an effect. So the effect was limited to those states that had Medicaid expansion in place. When we turn to the right, this is looking at individuals who had Medicaid and Medicare at time of transplant and not surprising, I think, is that you didn't see that there was an effect of being in an expansion state, but those individuals and similar results were seen actually when they looked at Medicare only. So the conclusion from this study was that there was an increase in transitions from private to Medicare, Medicaid programs in expansion states, suggesting this is an important safety net. Now I'm going to turn to the topic of immigrants, a special group of uninsured. Shown on the figure is a recognition that among the uninsured in the U.S., there's a significant proportion that are non-citizens. So among lawfully present immigrants, 25% are insured, and among undocumented immigrants, 46% are uninsured. And immigrants account for 27% of the uninsured population in the U.S., and you can see that two thirds of them reside in the eight states listed. And of the 20 million immigrants, 11 million are undocumented. What is the pathway for liver transplantation for immigrants? Well, I'm showing the pathway in California. There's also a similar pathway in New York and other states that have a permissive or an expanded Medicaid option for such patients. So first of all, they have to have pru-call, which is a status in which they're, regardless, permanently residing under color of law. This is specifically for non-citizens residing in the U.S. known to immigration service and with no, who have no intent of deporting. With pru-call, they can apply for Medicaid, and if they have such an option available in their state, they have to meet the usual requirements for Medicaid related to income and residency in the state. But in addition, there's a need to be a qualified, of qualified status, which means they have had to be an eligible immigrant for at least a five-year period. There's a few exceptions, refugees and some humanitarian immigrants who don't need this five-year requirement, but most do. And then they're eligible typically for a managed Medicaid plan that would direct their liver care and indicate where the liver transplant will occur. And sometimes these are occurring in disparate areas and may be very different from where they're getting their primary care, making coordination of care especially important. There's a high variability in terms of whether patients can access that pathway I just showed you, because it requires the state to have an access to Medicaid for immigrants. So in California, we indeed have that, qualified immigrants meeting the five-year requirement and other things, those with Procol and others are eligible for state Medicaid. But I've shown you two other states, for example, where this is not the case. And if we look across the U.S., state by state, there's great variability in terms of whether immigrants are eligible for the Medicaid programs. So not surprising with some of these challenges in access in terms of whether a Medicaid is available, there's relatively infrequent transportation of immigrants. This is data from March 2012 to 2018, 166 were transplanted, 0.4% of the total transplants. Compare that to the 297 transplant tourists that were transplanted during the same time. Shown on the table is, first of all, the proportion by state. And you can see that California and New York account for about 60% of the unauthorized immigrant liver transplants that are performed during this period of time. And it aligns with the state's sort of permissive rules regarding Medicaid access for patients. They do well. Shown here are the outcomes of the patients. While they are younger, they tend to have higher male scores and more frequent need for renal replacement therapy. But as you can see here, graft and patient survival are excellent and comparable to residents in terms of outcomes. There's been another development in 2020 that may further reduce the immigrants' access to liver transplantation. And that's the final rule on public charges, a rule in which now immigrants who utilize social programs, including non-emergency Medicaid, that they may be viewed as a public charge. And that may have negative consequences in terms of their consideration under their immigration request. Before the social programs were considered a public charge, only 3% of undocumented immigrants would have met the public charge rule. But under this revised definition in which any use of social programs would be counted, as many as 47% of current undocumented immigrants might meet the public charge rule. And so even lawfully present immigrants and persons living in mixed status families may choose to not participate in social programs like Medicaid due to concerns regarding how that may impact on their future immigration goals. So this is a very important development. All of this is to say that liver transplantation for the uninsured is complex. There's a network of programs and partnerships they need to be aware of. They all have distinct eligibility criteria. Enrollment procedures can be complex. Certainly for undocumented immigrants, not as citizens, they have to also know what the rules are by their region, how it might change over time, both at the state and the federal level. All of this is to say that they need help. And I think for programs committed to assisting the uninsured, that there's an important role for health care advocates and navigators to assist the patients in gaining access to regional options. So to summarize, I've shown you that the underinsured patients have reduced wait lists and post-liver transplant survival. It's not new. But what is new is that we're seeing an increased proportion of these patients over time. So I think this points to the importance of research to determine the factors that underlie this disparity between those with private versus Medicaid, Medicare insurance. ACA and Medicaid expansion have been beneficial. They've led to more access to liver transplant, although it's a little uneven by racial, ethnic groups. We're seeing increasing proportion of patients post-transplant transitioning from private to public insurance, highlighting the importance of this as a safety net. But I think the full impact of Medicaid expansion is not yet realized since many states have more recently come on board. And I think we'll see even more evidence of its potential benefits in the years ahead. And then finally, immigrants. This is a group in which there's still so much to be done. Limited access to transplantation, geographic disparities largely related to state policies and access to Medicaid. The new final rule on public charges poses an additional barrier. And for this reason, I think it's important that we have experts to assist the patients in navigating access, both immigrants as well as other uninsured groups. And with that, I'll close. I want to thank you very much for your attention.

liver transplantation disparities

gender disparities

pediatric liver transplantation

race disparities

socioeconomic status disparities

disease severity recognition

referrals for liver transplantation

MELD score calculations

organ allocation policies

post-transplant outcomes

Medicaid expansion effects