National Academy of Medicine

Keynote Address: Assessing Progress in End-of-Life and Serious Illness Care

Patrick Conway

Deputy Administrator for Innovation and Quality & Chief Medical Officer
Centers for Medicare & Medicaid Services

BACKGROUND

“Assessing Progress in End-of-Life and Serious Illness Care” was a private working event for representatives from more than 90 multidisciplinary stakeholder organizations hosted by the National Academy of Medicine on May 23, 2016. What follows is an edited transcript of the keynote address delivered by Patrick Conway. 

Dr. Patrick Conway

Dr. Patrick Conway

I want to thank first the National Academy of Medicine, and, looking around the room, I want to thank all of you for the work that you do in end-of-life care, palliative care, and generally across the United States. There are a few of things I’d like to call out. One: the report [Dying in America] is incredibly well done. Two: I think it is having a major impact across America—and I want to recognize Victor Dzau [president of the National Academy of Medicine], the report committee, and all of you for focusing on the implementation, because the implementation phase is when these reports really make a difference. Three: I think we’ve still got work to do, together, collectively, to improve. So our journey is far from over.

I’m going to weave in a couple patient stories, including some that are personal, and try to hit on five major points along the way. So, to begin, my father had cancer for many years. He passed away on December 5, 2007, and had a good death. We happen to have a family of four children—three doctors, one person who’s in the business of health care—and the process required a lot of intervention from us personally, including, at one point, me flying to talk to a hospital medicine colleague. I said, my father has been clear about where he wants to die, which is at home. I recognize that you’re worried about his blood pressure number and his potassium level, but let me be clear that he’s been clear about his wishes, as has our family. We were able to take him home from the hospital. This was a well-meaning physician. Like most care teams, there was no mal-intent. It was someone trying to take good care of my father.

Unfortunately, we still struggle with patient-centered care delivery. My father-in-law just passed away and had similar experiences. My wife was there over 50 percent of the days for a number of months. That taught me—we have four children—it is hard to be a single parent. I was only a single parent for weeks, but I can tell you it taught me to appreciate my wife even more deeply.

Now I want to sort of pivot and reflect. I’ve been at CMS [Centers for Medicare & Medicaid Services] for more than 5 years. I’m now the longest-serving chief medical officer in CMS history. I don’t know what that says about me or the job.

I had the honor to work with [U.S. Department of Health and Human Services] Secretary [Kathleen] Sebelius, who was a wonderful leader. I think we’ve gone from the “death panel” rhetoric at one period to hopefully a different period where we have more reasonable discussions about these issues. I will say, though, for the [Medicare] Part B drug model that we’re implementing, we cited an evidence-based report on paying for value in drugs—and, within a matter of days, people said, that report has quality-adjusted life years, therefore, CMS is reinstituting the death panels in the drug arena. So, unfortunately, we haven’t totally moved past that rhetoric in many ways. This is just a proposed model, and we are learning, and we are taking feedback, and we will make adjustments as needed. But I think it doesn’t help the discussion when we dive to non-evidence-based rhetoric.

I will hit on five areas in the end-of-life arena. One, on quality measures—over the last 5-plus years, we’ve [CMS has] implemented quality measures in the Medicare hospice arena for the first time. We think that was a big step forward. We also think there’s much more work to do. Everything from clinical quality measures to  patient and family experience-of-care measures are being implemented in the hospice and palliative care arena, [through] a Hospice Item Set.

We’re focusing on core, standard dataset collection across settings, including in the hospice and palliative care arena. We’re going to report these results publicly in hospice and palliative care. We think this sets us up for a much more patient-centered quality-of-care assessment in hospice and palliative care. We’ve seen in other programs that when we start measuring and reporting and ultimately paying for performance on quality measures, quality improves. We’ve seen this in the hospital and other settings.

Two—payment. In hospice, we also proposed changes where we’re paying higher for days 1 through 60 in hospice and also at the very end of life, when we know increased services need to be delivered. So we’re trying to learn and adjust payments to meet patient needs.

The Advance Care Planning payment is a huge step forward. I will say, and this relates to the rhetoric discussion, that there were challenges around proposing and finalizing that payment. We obviously thought it was the right thing to do and we did it. Now, our challenge is uptake and what’s really in that payment and what’s really occurring in those conversations, and whether it is as evidence-based and patient-and-family-centered as it could and should be. We will need to continue to improve the quality and impact of these discussions.

Three—quality improvement and some of our [CMS’s] new models of care. We’re increasingly seeing  Accountable Care Organizations and comprehensive primary care models focus on end of life and palliative care. For some organizations this was a big part of their work, but not all, so we are trying to spread this focus and learning. We have an affinity group in the Comprehensive Primary Care Initiative right now, which is one of our advanced primary care models, that’s focused on effective end-of-life care and  improving quality. We increasingly talk to our ACOs about all aspects of care delivery, including the importance of end-of-life care in the ACO population. There’s more work to do. I don’t want to overplay this. I’m not going to say that every ACO is providing highly reliable, high-quality end-of-life care, but we’re trying to work with our partners to move in the right direction.

Four—directly in the Innovation Center and new care models—let me give you a little more information about the Medicare Care Choices model. We launched in January 2016 a model through which we will be paying for hospice and palliative care services at the same time as curative care services. We have more than 140 hospice and palliative care organizations in two cohorts covering 39 states, so it’s a very large model. Evidence from the private market shows better quality of care, better patient and caregiver experience by far, lower cost, and longer longevity for concurrent palliative and curative services. So, we’re testing that in Medicare. If it works—and we think it will improve quality and lower costs—and if so, we could expand that nationally.

We’ll learn from the model, too. We chose a price point of $400 per beneficiary per month. One could argue it should be much more flexible price point based on the services delivered. We’re open to thinking about that. We selected certain eligibility criteria and recently adjusted them based on feedback that the criteria were actually too tight, limiting the uptake of the model. We’re going to learn as we implement this, so please keep that feedback coming. It helps us improve.

We think there will be 100,000 patients or more who eventually go through this model as we learn, and we think it’s a huge opportunity to change the paradigm. Congress could obviously act if they chose to change the law around hospice. But the reality of patient care is people aren’t getting curative care and then suddenly getting hospice and palliative care services—there’s a much more natural transition, and people should be able to elect how they go through that transition.

My own father had been cared for at MD Anderson [Cancer Center]. For his care team at the time, for going to hospice, he was going to have to withdraw from the clinical trial he was on, which for him as a scientist—he was a PhD chemist—was a big deal. Even though he realized it probably wasn’t going help him personally, he was very committed to the research process. So there are nuances and context here down to a really deep level where we still need to work to improve. I would argue, not because of my father, but for any patient—if they wanted to participate in a trial process, plus hospice and palliative care, we should enable that. Whether it’s research protocol rules or payment rules, we shouldn’t have those types of issues inhibit a patient and family’s choice.

Last, on the payment side—we don’t think Care Choices is the end of payment learning in the hospice and palliative care arena. Whether you call it an advanced illness model or another name, how do you pay for care before someone is hospice-eligible? And how do you have a payment model that supports people who are quite sick, frail, et cetera, and need a higher level of services? We’re going to need your help in defining that. We have to define it based on the evidence. And, unfortunately, although most providers do things for the right reasons, there’s a segment of our providers that aren’t always doing things for the right reasons. We have to figure out how we have criteria and the ability to monitor and make sure that we’re getting better patient care and better patient outcomes and not behavior that doesn’t lead to better patient care and better patient outcomes. So, help us on this advanced illness model.

I’ll end with a patient story. I am still a practicing physician in the hospital medicine setting, and this relates to education and training in the end-of-life arena. I don’t know the answer here, but I do think our education and training paradigm has left something to be desired. And I’ll expose my own deficiencies in education and training—I’m trying to model the learning model, if you will.

I practice at Children’s National Medical Center as a hospital medicine attending. When I was a resident in the pediatric ICU [intensive care unit], in the 30 days of my first rotation, my intern year, we had 11 children die, which is an incredibly high rate for a pediatric ICU. I learned a lot that I did not learn in med school about end-of-life discussions. I learned a lot about those discussions, from watching the attending physicians and fellows I was with, both the good and the bad. That impacted me in a really deep way.

I now take care of children, mainly with multiple chronic conditions. There was a mother—I won’t give details here, but she said I could tell this story broadly—I don’t very often. I’m on hospital medicine service just on weekends. I used to do it for full months and weeks at a time. One of my regrets is, in a month or a week, you can’t really get to know a family. On a weekend, it’s hard, because it’s two days, 48 hours. I met this mother and child multiple times, which shows they were in the hospital a lot. Her daughter had a terminal illness that everybody knew was terminal. Many, many interventions were being done again and again and again. She happens to also be a single mother and I’d gotten to know her well enough that we sat down and we were talking and I just opened up the discussion. And I read the report [Dying in America]—I may not have done it as well as described in the report, but I opened up to her that she and her family had choices. I just wanted her to know that she had choices about her and her child. That became a longer discussion—she actually got upset a bit, in the beginning, and then we got to a better place, even in the course of that weekend.

I saw her again, about a month later. And she was in a different place with her and her child. And then I saw her one more time, in a different place—she had made different choices around care. And then I saw her in CMS. This is a true story. She said, do you remember me? I said, of course I remember you. She said, I transferred to work with you. I said, thank you and tell me more. Actually I didn’t know for sure—because this is the sad part about hospital medicine—what had happened with her and her child. She said, you know, after those discussions, I did decide to take a different course with my child and we went into hospice and palliative care. She thanked me for the care provided to her and her child and our discussions.  We both broke down, to be honest, and cried. There’s probably many people that do this better than I, but it’s a tangible story of why these conversations matter.

So, with that, I will stop. I want to thank you again for the report. I want to thank all of you for being here today. I really appreciate your time. Thanks a lot.

Q&A Session

Can you touch a little more on the gaps in the training and education process, and where you might like to see that go?

I’ll comment, but I am not an expert on education and training. My guess—my hypothesis—is we need increased training across the paradigm of medical education and also across the paradigm of professions, the whole array of professional education. Because the evidence shows, and some of this is cited in the report, that interdisciplinary care teams matter immensely. The training of the different team members matters, and a high-functioning care team—which, as a quality improvement person who has studied context at one point in my life I know is hard to measure—the evidence says really matters in terms of outcomes. Therefore, I think training across those professions needs to be across the whole course. And, by the way, we’re going to have to do retraining of a lot of folks who— maybe including myself at one point, I hope I’m better than I was ten years ago—need training on how to deliver this care well, how to have [end-of-life] discussions, how to continue to improve over time.

Also, in some areas, whether it be in geriatrics or in the hospice and palliative care arena, we’ve just got a dearth in the number of people trained. So, we also have to think about how we fill that gap. And on the interdisciplinary part, just to be sure I’m explicit, I mean nursing, social work, pastoral care, et cetera, the entire interdisciplinary care team. I visit some places that have a true model of that type of care delivery. Unfortunately, that model is not reliably spread across the U.S. So the problem is how do you solve for that both at a macro level and at an individual/hospital/physician/clinician/clinic et cetera level.

Regarding the graduate medical education (GME) dollars that CMS is in charge of. We pay for slots for cardiologists, neurologists, and plastic surgeons, but there is not a single slot for palliative care. What are we going to do about that?

Great question. On the GME dollars, historically, the agency has interpreted that we have very minimal to almost no flexibility in how we allocate those dollars. So when various slots come up, we go through a process that, I’ll be honest, I find not optimal. I’m not making any comment about what should or should not be in the president’s budget. There was an IOM report on graduate medical education, which was very well done, but probably 90 percent of it would require a change in law for CMS to be able to do. We think the GME mechanisms are relatively strictly statutorily prescribed and do not have flexibility. That is the agency’s interpretation. We are always willing to explore flexibility administratively. Also, and I know there’s been some discussion on the Hill about this, if Congress were to either tell us to do, or give us funding to do pilots, we would do that. If Congress were to change the statutory formula, and we had more flexibility, we would take advantage of it, so we are very open in the GME arena for flexibility and then thinking about how we would use that flexibility.

Could you update us on the Independence at Home Demonstration project? Because I see that as such a key piece for delivering palliative care for those who aren’t quite hospice-eligible.

We’re incredibly excited about the results so far. We have very positive first-year results—very positive quality, positive patient and family experience, and more than $3,000 per beneficiary per year of savings. For a CMS model, that’s very large savings. Second-year results we’re working on now. There’s some detailed evaluation work that we’re going back and forth on with the participants. It’s actually hard to match this cohort—not to get too nitty gritty—they’re very sick people, and you want to do it well. But we think, depending on the second-year results, Independence at Home would have a chance for expansion. It is not an Innovation Center statutory model, so, therefore, it cannot be expanded administratively. So, the pathway for Independence at Home as it stands right now would be, we share those second-year results with you and Congress and everyone else, then they [Congress] would have to write a statute to expand. The good news is that if it improves quality and lowers costs, this is the kind of thing that gets bipartisan support. But we’ll need all of you to—I can never ask you to lobby—but to make sure Congress is educated. There are people on the Hill that are really interested in this model in a good way—they’re excited about it. We’re excited that they’re excited. So, we’ll try to get those second-year results out to potentially expand.

Even more broadly, on the caregiver issue, we had some Innovation Awards in the caregiver arena, including both caregivers and community health workers. We’ve tried to think about new population-based payments where people pay for caregiving and advanced primary care. You have to also come at it in other ways—is there anything in the core payment systems that we could do within the law to support caregiving in a more direct way? The amount of uncompensated caregiving in this country, by you and millions of others, is huge. So, is there a way to test this, and what would be the model for paying for caregiving more reliably, if that makes sense. Independence at Home is a key component but not the only component.

What helped your father have a good passing? Are there ways to take that model and expand?

He had a compassionate care team that was interdisciplinary and well-trained. I think he had relatively effective care coordination most of the time between primary care and specialty care, although we could imagine that could have been more well-coordinated, and that’s generally true across our health system. I think he happened to have a family, to be honest, with resources, knowledge, et cetera, which everyone doesn’t have. So that gets down to social determinants of health and how you invest in communities and how you invest in families and caregivers, all of which is important and is a serious long-term investment. Those are some of the things. I will say there were still opportunities for improvement, and that’s nothing against anyone on his care team. The macro message is how do we make our system as reliable as possible to deliver patient-and-family-centered care as close to every single time as possible?

How can entrepreneurship play a role in the end-of-life care arena?

We’re setting up various payment models, and we try to foster innovation and learning, including both external and peer-to-peer learning. We are seeing entrepreneurs entering this space in a pretty significant way, which we think is a positive direction. I met someone recently from a very large entrepreneur background—the exciting part is, he actually hadn’t been in health care with his investment dollars in a long time and now he is in a big way—and I think we’re seeing that more commonly. The innovation is around care coordination, instead of the next big tower or whatever else you could build with capital. There’s still work to do here, but I think we’re seeing entrepreneurial innovation around care delivery, and care delivery for frail, elderly, disabled populations, which we think is incredibly positive and we want to foster more of.