Welcome, everybody, to the on-demand session entitled, Value-Based Medicine in Hepatology. I'm Dr. Kimberly Brown from Henry Ford Hospital in Detroit. I'm Elliot Tapper from the University of Michigan in Ann Arbor. I think we have a really exciting program for you today, talking about healthcare delivery science as it applies to our patients and looking at the preservation of value from our clinical trials. We have a number of outstanding speakers. To top it off will be Dr. Marina Serper. She is the Assistant Professor at the University of Pennsylvania and the Associate Program Director for Research for the Gastroenterology Fellowship. Her first talk will be entitled, Doing It Right, Value-Based Healthcare Delivery Science. Our next talk will be delivered by Dr. Sumit Asrani, Associate Professor at Texas A&M and the Medical Director for the Center for Advanced Liver Disease at Baylor University. He'll be speaking to us on achieving value by preventing hospitalization. Our next speaker, Dr. Vishal Patel, Consultant Hepatologist at Kings College Hospital in England, will be giving us, I think, a great lecture on achieving high-value care through clinical trials. Our next talk, High-Value Care at the End of Life, will be given to us by Dr. Nneka Ufri, Transplant Hepatologist at the Mass General Hospital and Harvard Medical School. Then finally, Dr. Jean Edmond, Vice Chair for Surgery, Chief of Transplantation Services, and the Program Director for Transplant and Hepatobiliary Surgery Fellowship at New York, Presbyterian Columbia, will be giving us our final lecture entitled, High-Value Care in Transplantation. Thank you so much for the opportunity to speak at this value-based medicine and hepatology session, and I'm happy to kick off the session. I'm Assistant Professor of Medicine and GI and Hepatology at the University of Pennsylvania, and a health services researcher and epidemiologist focused on the study of delivering high-quality care in hepatology. Here are my disclosures. Since I'm the first speaker, we will first start by defining value-based care, talking about healthcare delivery science, discuss a little bit about where we want to be and where we are today, and then I'll show you some examples of care delivery models and interventions that have worked in hepatology and beyond. What is value? Well, value very simply is outcome divided by cost. Value in healthcare can be defined as the measured improvement in a patient's health outcomes for the cost of achieving that improvement. Value-based care essentially is better care, better health at a lower cost. In a perfect utopian system of value-based care, patients, providers, health insurance payers, suppliers of medications, devices, or drugs all benefit by having lower costs, higher efficiencies, and better outcomes, and overall society benefits with reduced overall healthcare spending and better overall health for the population. As you know, we're quite far from that utopian vision right now. More than a decade ago now, Porter and Teisberg came up with the six core steps of what a health system with value-based care is able to do. As I describe these, think about if your health system fits into any of these core principles. A health system should have integrated practice units, be able to measure costs and outcomes for every patient, start moving towards bundled payment system of care rather than fee-for-service, is able to integrate care delivery across facilities and the care continuum, can expand excellent care beyond just the reach of that particular center across a geographic distance, and has a suitable information technology platform. Really since the passage of the Affordable Care Act in 2010, Medicare and other private payers have implemented multiple value-based care payment programs, really focused on rewarding lower cost care with better or similar outcomes. A variety of programs have now been enacted and Medicare has really been at the forefront of this, covering 1.1 million clinicians in the United States. One of the programs that may be well-known to some is this merit-based incentive payment program, MIPS, which is now the default track for those who participate in Medicare, which in the United States is virtually all of us. There's also the advanced alternative payment model with accountable care organizations. We don't have time to delve into all the different mechanisms of how payers see value and reimburse or penalize value and quality. But the overall take, at least for MIPS and also other programs, is that results have been mixed. High performers, health systems that are large, well-resourced are able to modify coding and to take advantage of some of these payments. Also, there's concern about high-risk, under-insured, uninsured, poor patients with poor health being further disenfranchised and marginalized by some of these value-based payment programs. Here's an example of a study done by researchers at the University of Pennsylvania at the effect of bundle payments on healthcare quality. This compared more than 200 hospitals with bundle payments to 700 without. Basically, at three years, showed a 1.2 percent decrease in total spending on high-cost conditions such as acute MI, COPD, etc. The decrease in spending really came from decreased length of stay at acute skilled nursing facilities. This was somewhat balanced by increased outpatient spending but still, there was overall a cost benefit and a net neutral effect on mortality. One could argue that this bundle payment program was modestly effective but by the same token, really not highly affected. We're really not where we want to be. Now shifting from the payer perspective, let's shift to the healthcare perspective and talk about delivery science. Delivery science really seeks to study how to improve the delivery of care. Care that we know that is evidence-based, when it's integrated into current clinical care, we like to call that implementation science. Care that can improve quality locally, we have termed that quality improvement. But it's important to know that you may see some of these terms and that they're not exactly interchangeable, they're quite often complementary. Both types of research and implementation can generate high-quality data that's generalizable, both at the local and the national level. This is important to keep in mind. Delivery science interventions, there are a variety. There's now actually a journal devoted to healthcare delivery science. Really, these kinds of interventions are focused on redesigning current care practices, compare different innovations, and also implementing rapid pragmatic trials that have very broad inclusive criteria for patients to participate. I'll give you a couple of examples of such care delivery models and studies today. Let's launch into our first example. If I was giving this talk maybe three years ago, this would have sounded innovative. But currently, telemedicine and telehepatology have really become a routine part of many of our practices. But I want to talk a little bit about the promises and pitfalls of telehepatology as a first example. Prior to COVID, because of regulatory hurdles and lack of reimbursement, really the Veterans Affairs, which is our largest integrated health system, was at the forefront of delivering hepatology care via telemedicine. Studies have shown that with telemedicine, referral to transplantation was more efficient. There were fewer unnecessary referrals, and time to transplantation in a Richmond VA setting with a specialized telemedicine subspecialty care model called ScanEcho was decreased by almost two months. Telehealth really showed a lot of promise in improving efficiency in liver transplant evaluation, and also mortality shown in another VA study. We also did a study before COVID, which was boutique at this point, where we looked at 63 patients and evaluated a live video telehealth model from the University of Pennsylvania Hepatology to a GI practice about 60 miles away. The program was feasible. It was very highly rated by patients and referring providers. However, could not really continue beyond the research setting due to lack of reimbursement. Then came COVID, and with the emergency provisions, due to COVID and the CARES Act, telemedicine use really skyrocketed. Here I'm showing you data generated by our team from the VA. You can see that telemedicine really took off. Although the visit volume was not at pre-COVID levels, telemedicine replaced most of the visits in the VA for a short period of time. This happened in a lot of healthcare settings. But what we have to keep in mind is that telehealth is not a panacea. A study out of UPenn showed that even though telehealth was accessible to patients, it was still less likely to occur among those who were older, and those who had limited English proficiency and lower income. Even though patients who were Latinx, Black, and had higher comorbidities were more likely to do telehealth, they were less likely to do video visits than phone visits. The quality of these visits could be quite variable. They're persistent barriers to efficient high-quality integration of telehealth. Then another survey study that we conducted in Chicago, we actually showed that although most older adults in the Chicago area were able to have telehealth visits, those with lower marginal health literacy were less likely to have video visits, less likely to recommend telehealth to someone else, less likely to understand instructions during the visit. Results were similar for those with limited English proficiency and with those with higher comorbidities. Although telehealth has a lot of promise, we still have to work to implement it efficiently into our clinical practices. Some of the best practices for telehealth that we will touch upon here just a little bit are to understand that we really need to meet patients halfway, train them on the technology, ideally have a direct link for patients to be able to connect without an app download and a requirement for a password, to also being able to enable multi-party connections with caregivers or others, and have interpreter services available for patients. Next, I'm going to switch to another model that I will spend a couple of minutes of describing. This is an interesting population health model from my health system and an approach for hepatitis C screening among a birth cohort of adults. This was a study conducted by my colleague in gastroenterology and our Center for Innovation, Dr. Siobhan Mehta. He implemented a pragmatic trial leveraging our University of Pennsylvania health system primary care practices. This study enrolled birth cohort adults, and it was actually two studies in one. The first study randomized patients eligible for hep C screening who hadn't had it in the past to receive a letter in the mail to talk to their doctor about the screening versus a lab order plus the letter. The second study examined the effect of certain motivational messages and also an online patient portal versus mail letters, and the study was funded by the health system. What we saw was very interesting that in the letter only group, about 20 percent of patients had hepatitis C screening at four months. However, this was more than twice the amount, 43 percent among those patients who received a lab order in addition to the letter telling them to screen. This opt-out approach was actually highly effective and similar at 12 months and by sex, race, and income categories. It was highly cost-effective, only 71 cents per patient in terms of a cost to be able to screen eligible patients for hepatitis C, so highly effective and highly cost-effective as well. You're welcome to read this study. The effects of portal versus mail letters were modest where a mail letter was a little bit more effective and behavioral messaging compared to usual messaging really had no effect. The last model I want to share with you is a behavioral intervention using remote patient monitoring. This is a very interesting study that is very appropriate in the context of the fact that our patients are increasingly using wearable devices. Even though we're starting to get data in hepatology about wearables, that data are still pretty sparse. There was recently a study published of an intervention of remote monitoring after liver transplant that did show some efficacy with remote monitoring of blood pressure and oxygen as well as temperature in reducing 90-day readmissions, but we're still developing the data. Dr. Duarte-Rojo has developed our LFIT exercise app for prehabilitation and liver transplantation. But the question is, how can we now use the technology that we have with the wearables, with a liver-specific app, and integrate this into clinical practice? In the last minute or so, I want to share with you a study conducted at UPenn among 500 adults at high risk for coronary disease in low-income neighborhoods in Philadelphia. This was a five-arm study where they actually had patients wear Fitbit devices, and the control arm received messages every day about their daily step counts. The other four arms were randomized to test the effect of gamification, which is giving patients alerts and badges and messages about them achieving or not achieving their goals, as well as predetermined assigned goals, as well as gradual versus immediate goals. I don't have time to go through the entire study, but I do want to let you know that the group that had the most effect was the one where patients received daily messages with tailored goals and gamification, but these goals were actually picked by the patient and assigned immediately without a gradual increase. In this arm, actually, patients on average were able to walk almost 1,400 more steps a day on average and the effect was sustained at eight weeks. Very powerful study that could be implemented in AFLD maybe in the prehab population, has a lot of promise. I'm going to summarize here and conclude by saying that the field of value-based care has certainly evolved and matured. From a payer perspective, results have really been quite modest to mixed in terms of our ability to get towards value-based care. However, delivery science and care redesign, pragmatic trials offer a lot of promise in how we deliver care in hepatology. We have seen several health system examples of improving access and deploying these pragmatic care delivery models and pragmatic trials. There's a lot more work to be done in this area for sure, but I think the field at this point is wide open, and we can definitely borrow a lot of research from other disciplines and heart failure, the primary care literature, and I look forward to seeing what we're going to be doing in the future in hepatology. I thank you for your attention. All right. Thank you, Dr. Brown and Dr. Tapper for this kind invitation to continue the discussion on value-based medicine in hepatology. What I would like to concentrate on is discuss achieving value by reducing hospitalization. My name is Sumit Asrani. I'm the Medical Director for the Center for Advanced Liver Disease at Baylor University Medical Center in Dallas, Texas. I have no disclosures. The two themes that I wanted to get across is one, to talk about the principles of value-based care as it relates to inpatient burden, and two, to describe some of these interventions across the care continuum to reduce avoidable hospitalizations. Now, to build on what Dr. Serpa mentioned, when we talk about high-value care, what we mean is high-quality health care outcomes at the lowest cost or lower cost. High-quality care is appropriate. It's timely. It improves patient outcomes. The cost we want to lower by minimizing unnecessary and duplicative care as well as avoiding unnecessary hospitalizations. We as providers and care team members take care of the patients across this continuum, from taking care of patients with cirrhosis, decompensation, leading them towards transplant, managing their liver cancer, and assisting them if they encounter premature mortality. But a bulk of our time is spent on management of decompensation, and we have our patients who utilize the ED a lot, have the admissions and have extensive tests. And we have about 10% of the patient population that we see that is responsible for significant healthcare utilization. And so the frustration both as a physician as well as a patient is sort of this hopscotch of episodic and reactive care where our patients are jumping from the outpatient to the ER, from ER to inpatient, back to the ER, never making it to the outpatient and having poor transition from the inpatient to outpatient setting. So how does one avoid this episodic and reactive care which tends to happen in our current system? And hence, I think to really achieve value, to reduce hospitalizations, one really needs to look at all these four domains, the outpatient aspect, ED aspect, inpatient aspect, and in the final transition from an inpatient to outpatient setting. And so let's start off with outpatient. I think the theme here is really developing chronic disease management models. Some of these include procedure clinics. And I think the big need in our clinics, as you know, it's sort of having access to extenders, so these are educators, nutrition, social work trainers, and really to educate our caregivers to also realize when is the inflection point to improve the urgency of what is needed for our patients. And this, some of these concepts that I just mentioned, sort of fall into this concept of sort of developing a subspecialty-centered medical home where the hepatologist, as well as the hepatologist team, acts sort of like the quarterback. We're managing the initial referral, really getting to know our patients, developing programs that are patient-centered, offering this continuity of care, care coordination, and also transitions, looking at improving our quality, all under the envelope of having a team-based care. Now, many of us, you know, listening to this talk already do this, but probably in a fragmented manner. And the question is, how do we coalesce this into a subspecialty-centered medical home? And this kind of chronic disease management model does work. If you look at a couple of studies I'm gonna highlight where the four components of their chronic disease management model was, one, care coordination, two, is decision support by using evidence-based protocol, three, is providing self-management support by education to patient as well as caregiver, and four, a comprehensive review of all data by highly utilizing their clinical information systems. Again, we do this in a piecemeal fashion anyway, but if we, again, combine this approach in terms of a chronic disease management model, is it associated with improved outcomes? And in this controlled study, receipt of a patient going through this paradigm of chronic disease management model was associated with lower risk of admissions as well as lower risk of readmission. And if you look at more long-term data, even survival, if you look at the figure on the right, the dark line up is improved survival with chronic disease management as compared to standard of care. And in adjusted models, receiving the standard of care was associated with a higher risk of mortality as compared to receiving the chronic disease management model. Building on this, our patients do end up in the emergency department, and the theme there is really focusing on liver-specific protocols, triage, as well as having a mechanism to assist with low acuity procedures. And the three points are having an easy button, eating point of care, as well as having access to procedures in a semi-urgent fashion. What is really needed for our patients is real-time outpatient appointment and triage, where we have the ability for our ER physician colleagues to phone a friend or have an easy button where patients, whether they come in at 2 p.m. or 2 a.m., have access to the infrastructure to set up, let's say, next day appointments or having access to outpatient procedures. Care coordination also works if we do it in the ER setting or the day hospital. So here's a paper that I think many of us are familiar with is the concept of a care management program where you have a patient coming to the ED with a sirus, having a concept of a day hospital. They get their endoscopy, ultrasound, paracentesis as needed, having a care management check-in with labs, education, treatment, having the ability to set up follow-up, and again, offering repeat care management check-ins as needed if the primary team is not available, as well as offering communication with the PCP. Now, this whole thing would probably take about five to nine hours in the pilot that they did, but this kind of care management protocol is associated with lower 30-day readmissions, your utilizations, lower days for hospital stay, lower risk of death, as well as lower cost per encounter. So this five to nine hour program does work if we have the infrastructure for it. What about smaller subsets or slices of this concept? So here's a pilot program by our Penn colleagues that I've also highlighted below one of the interventions by our colleagues at Cornell is if you have patients with cirrhosis with a thyrus in the ER or if, let's say, the hepatologist of record has an outpatient who really needs it urgently, having this four-pronged approach in this pilot study, first part being an earlier triage from the ED to an ED observation unit, two, having priority with the procedure team for consultation and procedure, three, having labs, supplies, and albumin on standby, and then having, fourth, a mechanism to follow up on results in the ED and or having GI reverse loss backup. This four-pronged approach in this pilot study was associated with lack of admissions needed for any of these patients that went through this, again, low-acute procedures, decreased ED stay, as well as more efficient outpatient follow-up as well as triage. So this kind of concept is possible with coordination with our ER colleagues. Flipping this to another thing that we encounter a lot is hepatic encephalopathy. Arun Jaswityan at Cornell co-chaired this group in collaboration with the American College of Emergency Physicians to develop sort of point-of-care tools for liver patients. This is one of them called ALTERED for management and protocol-driven care of patients that present with hepatic encephalopathy, again, providing some kind of tool as the ER is frontline for many of our patients and having sort of systematic treatment as well as triage for patients that come in with hepatic encephalopathy. Now, our patients do get admitted, as you know. I think in the inpatient aspect to really reduce this variation, I think what is needed is protocols, checklists, and really looking at some of these quality measures that have been vetted. So we, as part of the Practice Metrics Committee over the last two, three years, have been involved in development of quality measures in cirrhosis care. This was a two- to three-year iterative process where we started off with published measures, added practice guidelines, included an expert panel, included the patient voice, included process outcomes, clinical outcomes, and combined this with patient-reported outcomes. And out of about 120 or 140 such measures, ended up with about 29 high-yield measures. We then worked on specification of these quality measures and now currently are involved in piloting some of these measures across 10 sites using the ASLD-supported Cirrhosis Quality Collaborative. For those who don't know about it, I think this is a good resource to check out and going to be looking at implementation of some of these measures. And if we just concentrate on the inpatient-based quality measures, again, these have been vetted. These have been shown to improve outcomes. These include antibiotics in GI bleed, appropriate SPP therapy, appropriate timing of endoscopy, treatment of hepatic encephalopathy with lactulose, post-discharge medications for hepatic encephalopathy, as well as a visit within two to four weeks of discharge. And those of us on this conference who are in large academic centers, many of these things are pretty routine for us, but there are smaller systems that do not have the infrastructure to support all of this. And hence, looking and decreasing variation in sort of some of these standards of care and being able to measure it, I think is important to, again, increase the value of the care that we provide. One way to do it is inpatient protocols and using checklists as an example. So in this study by our Michigan colleagues, looking at patients, inpatient with decompensated cirrhosis, for those standard of care, was it those that underwent an initial manual checklist and then eventually an electronic decision support for two aspects of liver disease management. One is goal-directed lactulose and rifaximin for overt HE, and two is antibiotics prophylaxis for SPP. And standard of care, which includes electronic decision support, those that got the electronic decision support, there was 40 to 60% lower odds of 30-day readmissions and 1.34 fewer hospital days for HE. So even in a large academic center, there is significant room for improvement once we standardize the way we do it. And then finally, this transition from inpatient to outpatient setting, having these high-risk patients. I think the theme here is, can we develop high-risk clinics and having access to this and managing the transition of care? Because we know that early follow-up is associated with a 40% lower risk of death versus those that don't receive an early follow-up based on VA data. In this study at our center, we identified about 150 unique patients, and this was identified by the hospitalist currently in the hospital system, who we felt were high-risk for readmission and were high utilizers of inpatient healthcare. We invested in having a dedicated NP and MA only for the high-risk clinic, where the four components was, one, a warm handoff, so the outpatient team would come to the inpatient setting and talk to the patient. Two, the transitional care phone calls within the first one to two days to go over common management issues. Three, the visit within seven days. And finally, the sort of liberty to say, you know what, take as many visits as you want to take care of these patients in the outpatient setting, resulting in about four visits per patient. And with this study, what we saw is that if we look at each of the roles, ER per patient, inpatient visits per patient, inpatient days per patient, as well as imaging per patient, pre-transitional care, post within 90 days, and post overall, and overall, there was a decline of about 60% or so in these measures. And the charges, we estimate decreased from 8.3 million to 2.9 million. So to summarize, to really achieve value by reducing hospitalizations, as one may surmise, you sort of need interventions at each of these four different levels. In the outpatient setting, having chronic disease management models, which should incorporate some kind of access to an outpatient procedure clinic. In the ER setting, having point-of-care protocols as well as, let's say, cap clinics for low-acuity procedures, and maybe even high-acuity procedures and getting them early access. In the inpatient setting, having protocols in place and really honing down on the quality measures that have been associated with improving outcomes and to decrease variation. And finally, for this transitional care, really being able to incorporate and link the inpatient and outpatient aspects of their care. In summary, I think to really achieve value, we need to look both at patient factors, provider factors, and system factors. From the patient standpoint, I think being acceptable, accepting alternate care models, I think is important. From a provider standpoint, I think we need to start defining success with regards to both quality and cost and the willingness to serve and be rewarded as a quarterback for patient-centered care and be able to adhere to quality measurement guidelines. And finally, the system level is really defining the system-level goals. Is it willing to support upstream interventions and accept changes in revenue with decreased hospitalizations? Is it willing to design appropriate tools to measure and implement processes and build the infrastructure for healthcare delivery, which includes care coordination and transitional care? Thank you again for your time. And now I'll hand it over to the next speaker. Hello, everyone. Thank you very much to the organizing committee at ASLD for inviting me to provide a perspective on achieving high-value-based healthcare, in particular through clinical trials. My name is Vishal Patel. I'm a hepatologist based at King's College Hospital in London, as well as a principal investigator and ensign lecturer at the Institute of Hepatology, Foundation for Liver Research in London. These are my disclosures. So I wanted to cover in the next few minutes the following, specifically what is value-based healthcare, on which I will touch, given that my co-speakers will cover some of this, thinking about specifically through the lens of clinical trial delivery in terms of goals and key consideration states success in delivering, as well as some of the challenges, and to think around the paradigm of the current UK poor dietitian research portfolio in developing value for money and improving patient outcomes. So what is value-based healthcare, which I've abbreviated in subsequent slides as VBH? It's a model of delivery of healthcare where the provider is paid based on patient health outcomes. Providers are rewarded for an improvement in patient health, a reduction in incidence in the effects of chronic disease, very relevant to our patients' cirrhosis, and for them to ultimately live healthier lives, and doing this in an evidence-based way. And this is absolutely key, and the reason why clinical trials are so essential in this. The value in value-based healthcare really is derived from measuring those health outcomes against the cost of delivering them. And in this, we think about the concepts of cost-effectiveness, resource utilization, and health economics. The benefits in value-based healthcare extend not only to individual patients and their caregivers, but it affects everyone up to society as a whole. If we think about this from a trial perspective, reduction in illness is all around the clinical endpoints that we decide on looking at. Healthcare costs look into health economics from a trial perspective. Greater patient satisfaction can be gleaned by qualitative research. And something that I feel very strongly about is undertaking, where possible, the biological work that allows us to better understand disease pathogenesis, deliver mechanistic outcomes, and potentially uncover new therapeutic targets. And all of these can be delivered in a randomized controlled trial setting with careful thought and organization. So value-based healthcare clearly is beneficial at every level, right through from patient to provider to resources, supplies, and society. And achieving this in a clinical trial setting really still hinges on the delivery of randomized controlled trials, gold standard of which would be double-blinded and placebo-controlled. Design is everything in clinical trial delivery. And I'm gonna touch upon a few of these aspects. Sample selection must be appropriate to the hypothesis in order to ensure that the outcomes and the data are generalizable to the population. We need to recruit enough patients to deliver the endpoints. And this is required, not just for the satisfaction of the methodologists in terms of power calculations, but to ensure that we can detect those clinically important differences if they truly exist between the interventions. We must have robust ways of randomizing in terms of concealing the treatment group from the investigators, because we need to ensure that we eliminate bias and minimize confounding variables. And the other groups that we of course conceal from are the patients and the methodologists. Standard of care must be universal and equitable between both groups to ensure that if there are changes seen that these are down to the intervention to minimize confounders. And intention to treat analysis is generally the way these trials are designed on the basis that patients and their outcomes are analyzed based on the original treatment group. The analysis from a hypothesis perspective must be based around the research question that led to the trial. And therefore the a priori hypothesis is tested. And we must at all costs avoid trawling to find significant differences unless they were part of the original trial. Key considerations in delivering of trials starts with setup. Intellectual considerations are important because primary and secondary endpoints, methodology and expertise all drive the initial questions that are asked and therefore the plans around delivery. Garnering support will be absolutely key. And the reason for this is that fellow colleagues will need to be involved and they need to believe in the research question. It's important to have good infrastructure available to deliver and a publication policy often helps in terms of incentivizing colleagues. Patient public engagement is absolutely crucial because these stakeholders have a significant role to play in terms of design and delivery of trials. And we found this both at a personal and institutional level in the studies we've been involved in. Protocols are the backbone of everything we do in a trial and therefore they do need thought and the contributions of the multiple experts you've been involving. We have good guidance by way of consort to ensure that we can a design a good protocol but also then update it and report trial outcomes and protocols by definition are iterative and are constantly undergoing amendments and improvements. If we can't recruit the appropriate number of patients you can't deliver the endpoints and therefore you must find a way to foster a culture of screening and recruitment to actively support these trials. Research must be seen as part of course service delivery and it's important to embed the training and resources to enable this to happen. Data handling given a trial is absolutely key. Robust requirements must be in place to ensure that you can be consistent and to have standard operating procedures so that clinicians who often deviate from these things can be protected from themselves. Financial considerations of course drive everything we do and ultimately we have to recognise that to do high quality work it is costly because of the number of sites, participants, staff and of course the interventions we're looking at and ultimately logistics will underpin everything you do and therefore having clinical research networks, proper R&D departments and the right staff in place is absolutely key and of course thinking about how to manufacture and generate the IMP and placebo and the ability to obviously locate patients. One of the major challenges in the last 18 months has been the effects of COVID and we've certainly experienced this and I'll share a bit about this with you but also what can be done. So looking at cirrhosis and obviously variceal hemorrhage the natural history of disease in arthrotics is to develop hypertension and then varices in the esophagus which increase in size and bleed. It is unclear at the moment whether patients with small varices bleed significantly given the data sets are poor but it is clear when varices progress that risk of bleeding increases and therefore it makes sense to think about intervening to prevent this. Non-sensitive bleach blockers have a role but it is unclear what this might be in preventing hemorrhage in those with grade 1 or small varices and of course in patients with medium to large varices where evidence suggests that we must intervene it remains unclear whether or not pharmacotherapy or band ligation remains the optimal choice. This is the reason that in the UK we have the BOP and Calibre trials running which will hopefully be able to answer some of these questions in a preventative manner. The National Institute for Health Research in the UK commissioned these particular trials on the basis that these are important questions that remain unanswered and the questions were driven by the UK hepatology community being able to engage with an IHR to convince them to provide the funding and resources required to answer these questions hopefully definitively once and for all. You can see from this slide that these two trials will be looking to recruit around 4,000 patients in total and therefore constitute the largest portal hypertension studies running globally at the moment and they are big studies and studies that will involve lots of different sites given the number of patients. What's nice about these is they're complementary and synergistic. The rationale for BOP is very much based on the fact that we know that prevention is key but the guidance certainly UK from 2015 and 2016 from BSG and NICE respectively suggests that there remains equipoise and uncertainty about whether intervening in patients with small varices to prevent primary haemorrhage is the right thing to do. This led to the Health Psychology Assessment Programme call which we answered and we were very pleased to be awarded the money to undertake the BOP trial. So the purpose of the trial is to understand whether or not carbidol reduces the rate of bleeding with a clinical and health economic end point. The primary objectives are really a co-primary as mentioned and the secondary objectives are as you would imagine from a clinical perspective. I'll elaborate a bit more on these in the next slides. The design is a traditional conventional multi-centre phase 4 triple bind and randomised controlled trial. We're looking at 1,002 anasartic patients and we will now be recruiting across 50 UK sites on the basis that the original 20 will not be able to deliver mostly due to the effect of the pandemic on the research capability at an individual site and hospital level. The funding allocated to this particular trial is 2.2 million and the endpoints as mentioned before are clinical as well as health economic from a primary perspective and the health economic is really key because unless we look at this we will not understand whether or not we're providing value for money for the intervention being purported. Secondary endpoints 1 to 6 are clinical and as one would expect in this sort of a trial but I really want to draw your attention to 7 to 10 which are around those patient factors, number 8 being health and resource utilisation again driving further health economic analyses, qualitative research to really understand the patient and primary caregiver experience and then something that we feel quite strongly about at King's which is the mechanistic sub-study MBOP to investigate pathophysiological pathways and hopefully try and look for new therapeutic targets and on this front we know that the paradigm of disease in cirrhosis is one which is generally progressive from injury in the liver to fibrosis and cirrhosis. The gut is incredibly important and is an area of interest for many of us but we have a poor understanding even this day and age around how the gut interacts with the liver is by way of the gut-liver axis. We know that the gut provides a barrier with different components which are outlined here but ultimately it is back to a translocation that drives a lot of the decompensating features in cirrhosis and preventing this has turned into a bit of a holy grail in some senses. So BOP will be able to understand whether or not we can prevent bleeding and whether non-selective beta blockers have a role here but of course all-cause decompensation thinking about the other complications our patients experience may also be impacted upon by COVID-19 and this is how MBOP was born with the thought that we can generate a hypothesis to test whether or not the intervention reduces bacterial translocation and therefore reduces all-cause decompensation and we can perform certain assays based on biological sampling to determine this. The flowchart for BOP is shown here and you can see that we plan to recruit just shy of 4,000 patients UK-wide excluding around two-thirds of those and the trial is very pragmatically designed such that after six weeks of involvement it's very much based around standard of care which has been seen every six months for clinical, biochemical and HCC surveillance in terms of assessment and annual endoscopies. MBOP has been designed pragmatically to mirror BOP and we plan to take annual samples on the back of the various visits that the patients will be undertaking when they see us for BOP. The key progress dates outlined here in terms of when we had the grant activated six months later achieving our first patient after getting regulatory approvals, site openings occurring at some rate now and to date we've recruited around 160 patients despite the pandemic. We however have had to pause for around 15 months and this has had a huge impact on progress as you can imagine and I think it's worth just spending a bit of time thinking about the impact of COVID-19 because value-based healthcare if it's depending on trials will be impacted by this particular issue and you can see here that screening which requires gastroscopies as an aerosol-generated procedure had to be paused in the middle of the pandemic understandably and if you can't screen you can't recruit and of course all the study-specific procedures, site initiation visits, monitoring and therefore study windows where we had protocol deviations were also affected. I'm very proud of the team at King's who were then able to put into place an action plan which had two primary aims. The first was to protect participant safety and the second to protect the primary endpoint and we put in a variety of measures which successfully mitigated for many of these. The BOP team is shown here and this is a really key slide for me because it kind of backs up and reinforces the point around collaboration and working together to deliver this sort of research and to summarize the ultimate goal of any therapy to research is to enable value-based healthcare that's front and center well-designed trials need to be methodologically sound to provide this evidence they can be incredibly powerful in terms of making these changes and we must consider what else we can do in addition to clinical endpoints and take advantage of that. We need real collaboration, engagement and input in terms of our experts. We must absolutely foster in the next generation of clinicians the need for them to actually participate and we need to break down barriers both at an individual organization structural level to enable this to happen. Thank you so much for your time and for participating today. Hi my name is Nneka Oufre and I'm a transplant hepatologist at Massachusetts General Hospital and I would like to thank ASLD for inviting me to give this talk on high-value palliative care and hepatology. These are my disclosures. The objectives of my talk are as follows to understand the current state of palliative care and specifically the palliative care needs of patients with advanced liver disease to understand the potential value of palliative care for patients with advanced liver disease and to define high-value palliative care and hepatology. So the prognosis of patients with advanced liver disease is poor. Even though advanced liver disease affects less than one percent of the U.S. population it is a leading cause of mortality in the United States. So for patients age 25 to 64 it's amongst the fourth to sixth leading cause of death and there's a 65 percent increase in deaths from advanced liver disease from 1999 to 2016. So approximately 44,000 U.S. individuals died from advanced liver disease in 2019. Along with their poor prognosis patients with advanced liver disease have high symptom burden and poor quality of life. In this meta-analysis patients with advanced liver disease had high physical and psychological symptom burden compared to patients with cancer, COPD, heart failure, and end-stage renal disease. Additionally our patients with advanced liver disease have high health care utilization. In this study out of Canada the researchers in this study wanted to assess end-of-life care costs for patients with chronic life-limiting illnesses and patients with end-stage liver disease were found to have end-of-life care health care costs that far outpaced those of patients with renal disease, heart failure, heart disease, diabetes mellitus, and mental health disorders. Additionally our patients with advanced liver disease received late advanced care planning. In a survey study of almost 400 attending physician members of ASLD that our group coordinated we assessed the timing and setting of initial advanced care planning discussions for patients with advanced liver disease and we found that the majority of attending hepatologists were waiting to have conversations related to resuscitation status, preferred site of death, and hospice care when patients were acutely hospitalized or when death was clearly imminent. Notably 81% of attending hepatologists believed advanced care planning discussions occur too late for patients with advanced liver disease. So we know what the problem is. Our patients despite their poor prognosis have high symptom burden, poor quality of life, high health care utilization, and poor quality end-of-life care. So current health care delivery models are not effectively meeting the palliative care needs of our patients with advanced liver disease. So next let's try and understand the potential value of palliative care for our patients. So what is palliative care? Palliative care is specialized medical care for patients facing a serious illness that focuses on providing patients with relief from the symptoms and stress of that illness. The goal is to improve the quality of life for the patient and family and palliative care is appropriate at any age and at any stage in a serious illness and can be provided along with curative intent. Notably palliative care can be delivered by anyone within the health care system, a hepatologist, a specialty palliative care clinician, a social worker, a behavioral health therapist, and more. So palliative care is more than just end-of-life care but it also encompasses physical and psychological aspects of care, advanced care planning, social and cultural issues, and spiritual, religious, and existential issues. Notably palliative care has been found to improve multiple outcomes in patients with other serious illnesses including survival, health care utilization, symptom burden, and more for patients with advanced cancers, chronic kidney disease, as well as advanced heart failure. And palliative care is becoming to be seen as more accepted in terms of a care model that delivers high-value care for patients with advanced liver disease as has been noted in this recent clinical practice update from the AGA that was published this year. So there is growing evidence showing the benefit of specialty palliative care for patients with advanced liver disease in which palliative care consultations were associated with reduced health care costs and end-of-life for patients with advanced liver disease, lower 90-day hospital readmission rates, as well as improved symptoms and increased advanced care planning documentation. We also know from large observational studies looking at the association of palliative care and health care utilization that for patients with cirrhosis palliative care is associated with a reduction in emergency department visits, hospital admissions, intensive care unit admissions, and increases the likelihood of patients having a death at home. So how can we really define high-value palliative care in hepatology? So we know that value is defined as quality divided by costs. And as we think about how to develop interventions and models of care that are targeted to the palliative care needs of our patients, thinking about domain-specific outcomes that we're hoping to measure and improve for our patients and their families will be fundamental in defining high-value palliative care in hepatology. So interventions that reduce pain, fatigue, frailty, insomnia, and muscle cramps, that improve depression and anxiety, that increase our patients and their families' illness and prognostic understanding, that reduce caregiver burden and increase the delivery of culturally humble care, that improve coping and well-being, and that reduce burdensome and non-beneficial end-of-life care all have the potential of increasing quality. So quality which can be improved through improving symptom burden, quality of life, quality of end-of-life care. And one might argue that by getting our patients healthier and happier to transplantation, we may actually improve transplant outcomes, post-transplant survivorship, and potentially even survival. However, we also have to think upfront about the costs of intervention delivery to ensure that palliative care interventions that we develop actually have value. So tracking metrics such as healthcare utilization, including ED visits, hospital readmissions, and procedure burden, the cost of care, as well as the cost of a palliative care intervention. And importantly, the burden of a palliative care intervention on patients and caregivers will be critically important in determining these interventions' values. So for that last point, intervention that could be high-quality in improving symptom burden and quality of life for our patients could also potentially be burdensome by increasing their transportation burden, increasing their time burden. And in that setting, even if that intervention has high quality, it would have low value because of its paradoxical increase in patient and caregiver burden. Lastly, as we think about how we can develop palliative care interventions in hepatology, we have to think upfront about how to optimize efficiency and scalability so that our interventions can reach as many patients with advanced liver disease as possible who are in different care settings. So how can we use a framework to define high-value palliative care in hepatology? So there are a few questions that we can ask. And I have this article that was written by Kate Curtright, as well as Scott Halpern that was published a few years in Annals of Internal Medicine, provides great guidance. So a few questions we should ask as we start to build the state of the science of palliative hepatology. Who is most likely to benefit from interventions we develop? What palliative care interventions are most beneficial and effective? Where should patients receive palliative care? When should patients receive palliative care? And how can efficient and equitable palliative care be implemented? The who is highlighted in black because these are things we can start to do in our clinical practice. But the what, the where, the when, and the how are areas of major gaps in our research that we need to start to fill so that we can improve the state of the science of palliative hepatology and start to deliver high-value palliative care to our patients. So who really focuses on how we can identify patients and families most likely to benefit from palliative care? And things that we can start to do in our clinical practice would be to start to collect patient-reported outcomes measures that assess a symptom burden and quality of life, the use of patient and caregiver-facing educational tools and investment in communication skills training to augment advanced care planning, screening for caregiver burden using validated indices, asking our patients in clinic what their sources of strength are and the role of religion and spirituality in their coping, and focusing on improving our communication skills training surrounding goals of care communication to improve our patients' end-of-life care. So really the focus here is to move away from a one-size-fits-all approach to palliative care interventions and to really start to target patients who have significant palliative care needs as well as their families so that we can start to identify places within our healthcare system in which they can get increased support, either increased support that we identify through our own hepatology clinics and our own clinical practice or referral to supportive care services such as specialty palliative care, mental health providers, behavioral health supports, chaplaincy, social work, and more. So what means, how can we actually identify the most effective components of palliative care and the most effective interventions? And here is where we really need an investment in improving the state of our science, specifically around symptom management through pharmacologic and non-pharmacologic clinical trials, the development and testing of patient and caregiver educational interventions to see if they help to improve illness understanding and health literacy, communication interventions to see if they help to improve outcomes related to advanced care planning, developing interventions that are targeted to caregivers and patients and caregivers as dyads, focusing on building psychosocial and behavioral interventions to focus on spirituality and coping, and then improving the state of our communication science to see if we can improve end-of-life care outcomes for our patients. Next is to try and determine when and where patients would receive the greatest benefits of getting palliative care. And as we're developing both primary and specialty palliative care interventions, identifying whether patients should receive those interventions in the home setting, in the clinical setting, or even in the inpatient setting will be of great importance as we try to assess the value of these interventions. Next and most importantly is the how. So how can we implement novel and effective palliative care delivery models to meet the needs of our patients? So there is an ongoing multi-site comparative effectiveness study called PalliVir that is looking at the role of specialty palliative care versus hepatologist-delivered palliative care, also known as primary palliative care, in improving care outcomes of patients with advanced liver disease and their caregivers. And in the context of a limited specialty palliative care workforce, this trial will be incredibly important in showing whether primary palliative care as delivered by hepatologists can actually lead to the delivery of high-value care for our patients. However, we know that palliative care can be delivered by individuals outside of hepatology and palliative care specialists. And thinking about the role of other individuals within our healthcare system to support patients and families and address their palliative care needs is an important area of additional science that we need to assess as we're developing high-value palliative care interventions for our patients. Furthermore, we need to think upfront about the scalability of our models of care to see whether care models should be delivered in person versus through telehealth, telephonic interventions, app-based interventions, communication-based interventions, and even the role of paper interventions as ways of educating patients and caregivers. So my key takeaways are as follows. Patients with advanced liver disease have substantial unmet palliative care needs. High-quality hepatology care is high-value palliative care. I recommend the routine screening for patient and caregiver reported outcomes to identify those with substantial palliative care needs and coordinate pathways for referral to local supportive care services. And there is an urgent need to improve the state-of-the-science symptom management, advanced care planning, goals of care communication, and palliative care delivery models. Lastly, as we try and focus on value, high-value palliative care interventions that we start to deliver for our patients with advanced liver disease should focus on issues related to efficiency, cost-effectiveness, and they should be targeted to the needs of our patients, and they should also be scalable. And with that, I would like to thank you for inviting me to this presentation. Good morning. I'm John Emo. I'm chief of transplant at Columbia, New York Presbyterian Hospital. I wanted to thank the association and the leadership of this group, Drs. Brown and Tapper, for the privilege of presenting our ideas about value care in liver transplantation. I have nothing to disclose. The traditional business school definition of value is the outcome divided by cost. Some writers might substitute quality, but outcome is the more specific concern, and cost is the consideration we have to challenge with. The challenges of this concept is that health care and liver transplantation, in particular, is an imperfect market. Patients, by and large, expect unlimited effort. The expense is distributed across payers. Considerations of cost will impact innovation and perhaps increase risk aversion for donor and patient selection, and I'll show you in a bit. If outcomes approach an equal level for all teams, then cost becomes the only discriminator, and perhaps that's a good thing. In the United States, we are not among the world leaders in terms of value health care. Bob Marion and his team at Arbor Research studied kidney transplants and identified that long-term kidney graft survival is lowest in the United States when compared to UK and England, perhaps attesting to the overall organization of our health care delivery. In liver transplantation, quality is readily recognized, both short- and long-term survival of candidates on the list as well as recipients, the complication rate, which is not so well captured, resource consumption, and here's the value proposition, and patients suffering inequality of life, which is not well captured in any model. The way that health care systems and transplant teams are governed in terms of cost is that more and more payment models have moved towards a phases of care global payment. In transplant, typically the pre-transplant, transplant episode, and post-transplant one year are recognized as global payment periods, and these prices can be negotiated with a payer with appropriate triggers for inliers and outliers and so forth. It gets very complex. The biggest challenge to transplant recipients is the wait list. This Copley painting is a great metaphor for it, and the quality of transplant teams, particularly the hepatologist role is best discerned in this period. A more positive metaphor perhaps is the goal of guiding the sheep that we gather into our wait list safely home at the end of the day. The importance of pre-transplant survival was very well demonstrated in this paper from Jesse Schultz from Cleveland Clinic, demonstrating that the discrimination between the very best and very lowest post-kidney transplant survival is far better than what is observed on the wait list. Survival benefit, which is not that well captured as a value proposition, because the costs of managing cirrhosis are not always considered in the value proposition for liver transplantation, and yet they're critical. The ecosystem affects behavior and outcomes of transplant patients, the center, the patients, and the referring docs, the region of sharing of organs, whether it be geographic or defined, and finally, the public media and government have a great stake in transplantation, so we have been exceptionally regulated all along. The metrics for the ecosystem remain a subject of debate. Up till recently, the only metric is post-transplant one-year patient graft survival. This does not serve to identify the commitment of teams to identifying patients, to managing the wait list, to generating living and deceased donors. Again, we noted that transplant rates and wait list outcomes are well represented. Post-transplant survival and life years gain reintegration, that's a very difficult metric. The quality of the surgery is central to the outcomes of transplantation, and advanced imaging and other techniques have brought transplantation, both living and deceased, to a very high level of reliability. If you look at the outcomes, this is our team. Just as an example, we're a better-than-average but not perfect team, but you see our one-year graft survival is 93%. Most teams are converging on between 92% to 95%. Interestingly, our live donor outcomes are 100%, yet we're only a four-star on that metric. It's a little example of what happens when you converge on perfect outcomes, which means that other metrics of quality and value need to be discerned beyond survival. The risk adjustment that is done by the SRTR, as an example, is a long list of variables, but the power of the coefficients is extremely limited. If you have a one-year expectation of 8% graft loss, of which half of it occurs in the first month or so, you have a very narrow window to discriminate between quality programs. Frank Irwin and colleagues at the UOPTM, the United Healthcare Group that governs transplantation, is probably the most successful and advanced group from the insurance industry evaluating transplant, had identified some challenges in terms of paying for transplant and moving forward. One of the things they highlighted was the importance of preemptive live donor transplant in the kidney population as a huge value proposition. This goes for livers as well. Both living donors as well as extended criteria donors, when properly used, improve the value proposition by getting people transplanted early. The team, an international team led by Pierre Clavier in Zurich, had studied with a variety of teams this idea of a benchmark liver transplant, when everything goes well, low operative duration, blood loss, etc., and found that between 8% and 30% of transplants met these benchmarks. Now, whether that's good or not might reflect either the skill of the teams or perhaps their aggressiveness in selecting increased risk patients. The American Society of Transplant Surgeons partnered with the American College of Surgeons to create a transplant-specific quality initiative modeled on NISQIP, which was born in the VA system, to try and develop more granular outcome predictions for abdominal organ transplant. This is still not fully rolled out yet, but a variety of granular outcomes such as hepatic artery thrombosis, bile leaks, biliary strictures, these are the things that perhaps could advance our ability to discriminate between teams. Professor Michael Porter at Harvard has articulated value-based healthcare delivery strategic agenda, integrated practice units around patient medical conditions. Transplant has always been ahead of the curve in this sort of thing. We are still not great at measuring outcomes and costs for every patient, in part because frequently the doctors in the hospital are separate corporation and in theory cannot share data, but the bundled payments, the delivery systems, geographic reach, and IT are all things where transplant programs were ahead of the game. If we want to optimize care, we obviously need multidisciplinary team throughout the continuum of care. We are big fans of the hepatologist being engaged in post-transplant care from the day one as well, rather than separating surgical and medical management protocol, care delivery, expertise throughout the continuum, and an advanced provider-based system, which again, most transplant programs have been ahead. Now, one thing Porter pointed out was that volume enables value. Although there is a justification for smaller transplant programs, it is easier to achieve high value with increased volume due to rapidly accumulating experience, moving to optimizing the team, improved training, and innovation. There are limits to value-based care, limited investment for growth, increased cost of innovation, decreased tolerance for risk, and the challenge is defining the cost ecosystem. There are benefits of transplantation for early transplant, which may not be properly measured and could result in apparent increased costs that are inappropriately attributed to transplant programs. Overall, though, we know quality and value for that matter when we see it. Quality costs less. Variability is the biggest challenge within teams. It's easier to control. Variability across teams is remarkable and persistent. The elimination of these things, better training, better patient and donor selection, more living donors, faster surgery, that's a training consequence, and integrated care delivery throughout the system. I thank you all for the privilege of sharing these ideas. Well, that was a terrific set of talks, and it is my privilege to make some closing remarks, where I'd like to share with you some of the things that I am particularly excited about with respect to each of our speakers' lectures. I personally don't have any disclosures that are pertinent here, so I wanted to begin with Dr. Serper's terrific lecture and leave a comment that we really need to focus on improving the methods of quality improvement, that if we seek to universalize the specific practices, we have to think about how we interact with the clinician in front of the patient. Anytime we do something, we have to consider that user in our designs to make sure that they're able to do something for the better of their care within the context of care delivery, so it's not causing any additional effort or time. A focus on electronic interventions, and every time we seek to do something different, we need to evaluate our interventions to make sure that we're on the right track. Dr. Esrani helped us think through the ways that we can achieve value by preventing hospitalization, particularly focusing on preventing decompensation, early detection of decompensation, and linkage to care, and optimizing outpatient care by building things like day hospitals that are equipped with the tools and clinicians that can help provide optimal care for patients with cirrhosis that can avoid the emergency department. Dr. Vishal Patel is doing some very exciting work across the Atlantic, and he reminds us that generating evidence through clinical trials is the lifeblood of quality. We cannot know if what we are doing is helping improve the missions of value-based medicine unless we have the highest quality evidence behind that, and we need people like Dr. Patel out there doing the clinical trials that will generate that body of evidence, and in order to do so, we have to be like him and adopt a trial-focused ethos, that we have to, when we see patients, think about how we can enroll them into trials, and if trials are lacking, we need to begin developing within our own neighborhoods the partnerships between the patients, the providers, the payers, and the funders that will generate more interest and active engagement in the conduct of trials for patients with cirrhosis in the United States and elsewhere. Dr. Ufere gave us an excellent talk about maintaining high-value care at the end of life. While we can often focus on trying to reduce costs to improve value, I think it's worth remembering that if we hold true to our principles, that we focus on improving quality of life for those with symptomatic cirrhosis, and we do what we can to help them avoid hospitalization, if we focus on that element of the value equation, cost containment will come. Dr. Emond talked about high-value care in liver transplantation, and here we have to maintain an active practice of stewardship with continuous introspection and improvement of our highly scrutinized procedures and outcomes. We must work to promote equity so that all patients, regardless of where they're from, their race or creed, have equal access to this life-saving intervention, and that we need to begin focusing on a definition of value that focuses not only on graft survival or survival of the patient after one year, but also their ability to thrive with this life-saving gift. And with that, I'd like to thank you for your attention and your interest in this topic.

clinical trials

data capture

monitoring

protocols

regulations

training staff

engaging stakeholders

trial findings

medical service delivery

hepatology

liver transplantation