Churning Patients Through the End of Life: A Podcast with Joan Teno

Eric: Welcome to the GeriPal podcast. This is Eric Widera.

Alex: This is Alex Smith.

Eric: And Alex, who do we have with us today?

Alex: Today we have an extra super special guest.

Eric: Super, super, super special.

Alex: Super special. We have Joan Teno who is a Professor of Medicine. She’s in geriatrics and palliative care and she’s at Oregon Health Sciences University up in Portland. Welcome to the GeriPal podcast Joan.

Joan: You know, I’m delighted to be here with you guys today. I can’t think of a better place to be on a Friday afternoon.

Eric: So, we got some great topics to talk about today. But before we do, do you have a song for Alex to sing?

Joan: I think we gotta choose a song for our time, A Change is Going to Come, because a lot of us have been working hard to try to change the system to be both patient and family-centered and I think we’re finally starting to see that change.

Eric: Let’s get a little change is ready.

Alex: A Little Change is Gonna Come. [Singing]

Eric: Nice.

Joan: Yay! Excellent!

Eric: Change is coming because we have a brand new setup here with a new mixer so-

Alex: New mixer, we got new effects.

Eric: So if you hear a little echo in there that’s the real change that’s coming.

Alex: Some chorus. Change. Change in palliative care, changing the singing. We’ll see, yeah.

Eric: Joan, thanks for again joining us today. What’s spearheaded this discussion is you just released a paper in JAMA, which feels like an update to one that you had published several years ago about site of death, place of care, and health care transitions among US beneficiaries. We’ll have a link to the current article on our GeriPal blog. It took me a while to think of the word blog.

Alex: We do have a blog.

Eric: We do. We occasionally post on that. But maybe we can take a step back, what got you interested in this subject?

Joan: I think when I started off doing this, a lot of my early career, like maybe the first five years of my career, was the failed SUPPORT study. And so as a junior faculty, what do you do after you’re involved in a $28 million study, which has not one p value. And it’s really totally significant. You can say I’ve had it quit, or you can dig in and try to really conduct some research to describe what’s going on, what are the opportunities to improve.

Another avenue I took is creating measurement tools to try to have better measures to examine a person and family experience of end of life care or care for seriously ill. This paper is something I’ve been always fascinated by, is the changes with financial incentives and how it impacts the dying experience. And I have to admit, when I originally do the first paper in, 2013 it was, we were a little bit disturbed. We jokingly called it the Mor Mor Mor paper after our senior author Vince Mor.

Alex: The finding of that paper was that people are getting more of everything?

Joan: More of everything.

Eric: What I remember from that, it was a more churn, if I remember. The one word that came out to mind was churn when I read your paper.

Joan: Yeah. I’ve always come up with very interesting words to try to describe things. And I think the first paper I wrote on churning there was an editorial all about my decision to use the word churn. But that was quite natural for me. I grew up in rural Pennsylvania and churning butter was something we did and it was something I hated to do, and I just thought it was one of those futile things that you did. Why can’t we just go to the store and buy it?

Eric: If I remember the order correctly, people were looking at positive indicators. Oh, more people are using hospice, less people are dying in the hospital. Hey, isn’t that great? But if I’m remembering correctly, the epiphany after reading your article for me was at least, okay, site of death isn’t enough. Because while people are not dying in the hospital, more likely to die at home and more likely to use hospice. They’re using a lot of hospice just for three days or less.

They were using a lot of ICU and hospitalizations in the last 90 days before death. And a lot more transitions between different levels of care. Am I that remembering correctly?

Joan: Yeah. Actually, our tagline was more ICU, more hospitalizations and more late transitions that are burdensome to dying persons in their family. I think one of the things we want to make this point with this original article, that it’s not all about where you die, it’s about what happens in that utilization and that last month of life. And whether you’re receiving care in the setting that you want.

So I’ve always been very concerned about using site of death as a quality measure. One of those concerns are, you could have someone who you manage really well over that last 90 days of life and they had to go into a hospice inpatient unit for the last three days because you couldn’t control their pain at home.

I would argue that the 87 days before, where they were at home and where they wanted to be. That is really good care. You had to do change in the setting of care because of their need for aggressive pain management.

Alex: Mm-hmm. So that’s terrific lead into this current paper. And just to set us up for this current paper, can you remind us that old paper covered what period and this new paper is different. I think not only in the period cover, but also that you’ve included Medicare Advantage people for the first, maybe the first time.

Joan: Well, I’m probably not the first person to use the Medicare Advantage data. But the original paper was 2000 to 2009. And what was really concerning to us was we saw a continued upward trend in ICUs going from around 24.3 to 29.2. We saw people making transitions in the last three days of their life from about 10% to 14%.

This new paper, we decided we want to do two things. First, we wanted to take the data out to 2015 and see what an early glimpse of the Affordable Care Act was and how it was impacting into life care. And then second big trend, despite the fact the people who created the ACA, they thought Medicare Advantage was going to startly sort of stable and decrease. Medicare Advantage continues to grow and it continues to be a very important part. And if anything, when you look at the Medicare Advantage data, there’s some striking differences in terms of hospitalizations in the last 90 days of life, last 30 days of life, increased use of hospice. So, I think there may be some lessons for us to learn from some well-functioning Medicare Advantage programs.

Alex: Mm-hmm (affirmative). But before we get into the nitty gritty results, I wonder if you could say a little bit more about these Medicare Advantage data. I mean, so we work in the VA and we sometimes submit some billing type of codes because they ask us to, but not because we actually bill for it. So we don’t have a lot of incentive. So, let’s just say that sometimes they’re probably not as accurate as they might be, just hypothetically speaking on the air for posterity, recorded, as they might be, say if I was in fee for service program and my income actually relied upon me billing accurately.

Eric: I don’t know what Alex is talking about. I bill 100% accurately all the time.

Alex: All the time. All the time.

Eric: If you’re listening Federal Government.

Joan: So, if there’s any CMS spies out there they can attest that your billing’s perfect. I think you’re very right. There’s a concern with using the fee codes around palliative care that people don’t use them because they’re not tied to reimbursement. What happened in 2008 and 2009 was CMS started requiring informational bills to be submitted from patients who had Medicare Advantage. And they did that. The calculation of disproportionate share and other GME typed funding. And so hospitals did have an incentive to submit that bill and to accurately portray what was going on, because it was tied to some funding.

Now, I think what the difference is, that bill to my understanding doesn’t undergo the same data edits a fee-for-service bill would go. I’m very fortunate to work with some really wonderful colleagues at Brown, and they’ve done some looking into how well did these bills, if you used them, represent 30 day re-hospitalizations. And the positive predictive values is quite high. So, I was confident that we could use them with the appropriate knowledge limitations.

One, critical access hospitals don’t submit these bills. And if there’s a Medicare Advantage plan that takes care of nothing but MA patients in their hospitals, they’re not going to submit these bills. So, as we report an article for the most part, this gives you a pretty accurate representation of what’s going on in MA. MA has been that little bit of a black box, what’s happening within the life care and MA. We don’t know. So this gives us some early glimpses on it.

And also recall that, when you have MA you have to submit assessment data with the OASIS, and you must submit MDS data. So that gives you some glimpses on what’s going on. It’s kind of amazing when you think nearly a third of the decedents are covered by MA and previously we really didn’t have knowledge on what’s happening to those decedents, and what their quality of care is.

Alex: Yeah.

Eric: Great. And just for our audience, when we say MA we mean Medicare Advantage.

Joan: Absolutely.

Eric: So, Joan, well, what did you actually find in your paper?

Joan: So we found, there’s really two tracks to talk about. What’s happening on fee-for-service and what’s different among Medicare Advantage. So why don’t we take fee-for-service first. So, we see a continued decline in people dying in acute care hospital. Increase use of hospice to nearly half the decedents. And what got me excited about these findings was we saw burdensome patterns of care decreasing. So, people who spent less than three days of hospice decreased from 14.2% in 2009 to 10.8%. People having three or more hospitalizations the last 90 days in life decreased from 11.5 to 7.1. The other thing is transitions between a nursing home and hospital and hospital nursing home nearly had a 50% reduction.

So if you just take a look at that, it looks like we’re heading in the right direction. One thing that we didn’t see a budge in was, the use of ICU in the last 30 days of life. Now, is the glass half empty or is the glass half full? I have to admit I was pretty excited that ICUs wasn’t going up.

Eric: Yeah. That’s a big-

Joan: I initially wanted to say this was a positive finding, but wisely I was working for a wonderful set of editors at JAMA, who are really the best. And they said listen, stabilizing at 29% is probably not a good thing, so they really helped me think through the message. Because I was sort of thinking like that palliative care physician, “Well, at least it’s not increasing!”

Eric: That’s my thought about it.

Alex: And then let’s talk a little bit about the Medicare Advantage and what’s going on there.

Joan: Medicare Advantage, I think there’s some lessons for us to learn from Medicare Advantage and just this whole notion of population-based health. Those findings that were really striking to me is compared to fee-for-service Medicare Advantage were nearly 10% less likely to be in the hospital in the last 90 days of life.

Alex: Yeah.

Joan: In the last 30 days of life, it went from 53.9 to 44.6. And they had a slightly lower use of ICU, by about two percentage point, was 27.4. They have probably one of the highest use of hospice and highest rate of people dying at home or the community. So I think, part of the million dollar question is, what is happening that’s causing these changes? And I think it’s really difficult to disentangle. A lot of people have been working really hard. And my hats off to the efforts to grow palliative care, The Conversation Project to promote advanced care planning. But for me, the one big key difference here also is the implementation of the Affordable Care Act and the 30 day re-hospitalization penalties.

I think that really caused hospitals to say, oh, wait a minute. I have to think about where these patients are going when they leave my doors, and how this might impact my bottom line. And I thought that provided a real ignitus for improving care at the close of life. So, I think it’s a combination of the things. You got to have the right processes and structures in place. But then the financial incentives aligning in the right direction really are helpful.

Alex: Right. And Joan, when we’re talking, just to be clear. When we’re talking about those financial incentives and penalties for readmission, does that impact both fee-for-service Medicare and Medicare Advantage? Is that really more on the fee-for-service side of things?

Joan: It’s more on the fee-for-service side. Although there are efforts towards really looking at what the 30 day re-hospitalizations are for MA, for Medicare Advantage.

Alex: Mm-hmm (affirmative).

Joan: I did some case studies as part of a program project grant that I was involved in. And I can remember going to these markets and talking to CFOs and chief transition officers, okay. Something I had not heard previously, a CTO, a chief transition officer, and they talked about they used to refer to something like 400 nursing homes and now they only refer to 40. I think there is some brilliance in CMS’s move to establish a penalty for 30 day re-hos. Because I think everybody would agree, is not just the hospital’s fault that’s going on here with these 30 day re-hospitalizations. If anything, when you have a lot of transitions going to nursing home, and those transitions get further out past weeks, probably is like huge role in the nursing home has to play in that re-hospitalization.

But CMS took a macro economic perspective, who has the money? The hospital. Who should we penalize and have them fix it? The hospital. And I think in a way what we’re seeing with some of these numbers is reflective of the fact that hospitals started to care about where that patient goes to after they leave their front door.

Alex: Mm-hmm (affirmative). You know Joan, I also want to pick up on this, speaking while we’re on the financial incentives thread. For Medicare Advantage, we have higher rates of hospice use and it looks like they’ve been higher over time. And I think one of my mentors, Ellen McCarthy, when I was a fellow, had published similar findings in JAMA. The question is, what is the story here? Is the story that Medicare Advantage is trying to save money because they can take patients off of their books and put them back into fee-for-service Medicare, where they’ll no longer be billed before these seriously ill, very sick patients who are likely to use a lot of services near the end of life? Or is it this is a great story here that Medicare Advantage is doing a terrific job at getting people who are seriously ill near the end of life into hospice?

Joan: So, I’m going to be that proverbial researcher and say we need more data. When have you not talked to someone who is funded and said but I need more money to study this. But I think we really need to look at the brief family members perceptions of the quality of care in MA versus fee-for-service.

I have a feeling what we’re going to find is that MA does a better job, and I’ll tell you why I have that hunch. I had a wonderful fellow from UCSF who did geriatrics at UW who published a great article in JAGS looking at the pattern of transitions in the last three days of life. Finding that, when you move someone in the last three days of life you have more unmet needs, more reported concerns and a lower rating in the quality of care. And the one thing that we saw that is a big reduction that’s occurring is in these transitions the last three days to life.

So, I have to believe that MA is doing a better job by either referring to hospice or providing oversight. Because when although it’s a carve out for the cost of that terminal illness, the MA plan is still responsible for the physician part of their care. So, my hunch, if I had to put my $5 down here, I would say that MA is probably doing a better job than fee-for-service because the incentives are aligning with the right care.

Eric: So who pays for hospice if they’re getting Medicare Advantage?

Joan: This is actually one of the very interesting policy debates that hasn’t been resolved yet. When someone is on Medicare Advantage, they carve out back into fee-for-service for the Medicare hospice benefit for all the care related to that terminal illness. But the MA plan still has some oversight on things that they were providing extra and also if your physician is in an MA plan, they still have oversight for that patient.

When we looked at this in the past, among those people who are live discharge from hospice, they go back to their MA plan. It’s not like they’re saying sayonara, goodbye. Like 99% of them go back to that MA plan. And we’ve published some other papers looking at life discharges and patterns of those life discharges and they’re very similar in MA and fee-for-service. As of the most recent data we have, we’re not seeing it. MAs have a very different pattern length of stay, or use of … They obviously are using hospice more. But we’re not seeing any of those things to suggest they’re inappropriately using hospice services.

Now, like anything, if you change the financial incentives that can always change. And I think, one thing you have to be very careful is we’re making this big transition of going from paying for more, to paying for value, but we don’t want paying for value to becoming paying for less care. So we’ve got to be careful in managing this transition. It’s got to be about quality. It can’t be about just saving money.

Eric: And I know earlier, when we were talking about why does Medicare Advantage look better in your data? You said that there’s some great people thinking about why this is and some great research, and we need more research around it. But just between you and me, and maybe some of our audience, what do you think it is?

Alex: The 12,000 listeners. Right.

Eric: Hypothetically.

Joan: So, hypothetically, you may want to ask, as a researcher, when you turn 65 in five years, are you going fee-for-service or Medicare Advantage? I will be registering with my local Medicare Advantage program which is Kaiser. I actually think they’re innovative and they’re doing a better job. And I think fee-for-service has incentives aligned with another hospitalization, another procedure, and not aligned with quality. And I think that’s why it’s so important that we have the right quality measures in place so that it’s about value, and value is not just saving money. What’s really important is meeting the needs of that seriously ill patient and their family.

And we’re in a foot race here, because the health care system, it’s out of the track and it’s changing. And we quickly need to get the right set of quality measures to ensure that patients are getting the care that they need, and that we’re not going to run into the problem of under-provision of care, which would be as equally terrible as over-providing care.

Alex: Mm-hmm (affirmative). Other messages from your research here?

Joan: Well, I think we really need to be very thankful for all the hard work that people are doing. Just look at the huge secular changes. First of all, the early innovators in hospice radically transformed end of life care in the United States, the second wave of big innovators for palliative care. I think it’s really important that we celebrate that, but we also remember that we’re not done yet.

I actually think where we are in our history is 1974 of the Medicare hospice benefit. And my prediction is when I qualify for Medicare in five years, there’s going to be a new alternative payment model that will help to provide access to palliative care and appropriate support services for people who are seriously ill. It’s really key in the next five to seven years. We’re going to shape that policy and then we’re going to be stuck with it for about 20 or 30 years. So, we need to work really hard and make sure we get it right. We don’t have a chance for failure.

My message is, let’s all really figure out how to do it, how to do it right. Make sure that the incentives align with providing high quality care and not just avoiding hospitalizations. Making sure that patients get the care that they need, and their families are supported during this very difficult time period.

Alex: Right. So it’s really a call to focus more on policy now, because what your paper is demonstrating is the impact of policy. Both policy in terms of hospice policy, policy in terms of incentivizing hospitals to prevent readmissions, which has also stimulated the growth of palliative care. What are the next steps that we should be working on in terms of policy?

Joan: Well, from my standpoint, I’m trying to work with RAND and a wonderful group of persons to come up with some measures that captures a seriously ill person’s perceptions of the quality of care so then we have actionable reports. And I think that’s something that we’re on our way with the CAHPS Hospice Survey, and we now need to extend that work into the seriously ill. I think one of the biggest things that we need to think about is how do we measure what we all say so important, this notion of goal of concordant care. I think it’s somewhat frustrating. We all know what it is and what we need to do, but how to measure it is still really complicated. Because you don’t want to introduce measures that says, “You must have resolution of your advanced care plans by the second visit to that palliative care physician, or that second visit to that family practitioner.”

Eric: People keep on trying that.

Alex: Trying that, exactly that. Yes.

Eric: It bugs me so much when I see that.

Joan: So we have some work to do. But, you know what’s really so exciting? Is, we have a workforce here that is in palliative care that wasn’t here 30 years ago. I swear, you could have put us all in the same ballroom, and if a bomb hit that ballroom you would have lost the entire hospice community, entire community of hospice and palliative care researchers and physicians.

We now have probably some of the most exciting investigators out there who are doing really good work. And I’m just very excited to watch where this field goes. It’s such a neat time period to be in. And this is really our chance to make change. We’ve arrived and we really need to, to really take advantage of this and now really head towards really trying to get the policy right.

The good news is there’s a lot of us who are working very hard to try to make sure we get the policy right. Like, anytime you’re dealing with policy, it’s gonna be some compromises, but you try to get as right as much as possible.

Alex: So, Joan, I got to ask, in our email exchange, when we’re trying to find a date and set up a topic and choose a song, you’d mention that you are considering titling a talk that you’re going to give at the upcoming annual Kathy Foley retreat, “Why we don’t need any more Joan Teno papers”.

Joan: Yeah.

Alex: I thought, “Oh, no. What are you saying? We need more Joan Teno papers.”

Joan: Actually, I’d like to argue you probably don’t. I actually have really tried to, where possible, become part of working with master contractors regarding policy in regards to CMS and the creation of alternative payment models and the evaluation of those alternative payment models. And to do that, you probably are gonna see my output of papers decrease, and that’s good. I actually think that’s what I should be doing at this time point in my life. You don’t need 292 from me. What we really need to try to do is spend the next five years really focusing on doing a really good evaluation of these various new models that we’re testing, and then think about how to implement them in such a way that they provide high quality care.

Eric: Well, I gotta say, another good part of that is that me and Ken Covinsky were joking that it would be nice if you saved room in JAMA and New England Journal for some other people.

Joan: Gladly. It will be gladly done. I should tell you that everything I do is part of a wonderful team effort. About 10 years ago, the team I work with, we do two things on every JAMA, New England Journal, or such paper. We have chocolate cake and a bottle of champagne. I’m very fortunate to have about nine bottles of champagne hanging up my house, and we’ve had a lot of really good chocolate cake. And most recently, with this paper coming out, I celebrated my 60th birthday with two of my co-authors having chocolate cake at a local restaurant in Seattle.

Eric: That’s wonderful.

Alex: That’s great.

Eric: So, for our audience, can you name two or three things that we should really be paying attention for from a policy perspective that’s going to impact either geriatrics or palliative care?

Joan: So, I think we need to be very focused on payment models and making sure they’re adequate to provide that care. We should be very focused on who’s in the denominator and who gets access to these services. And then third, we need to really make sure that there’s really actionable quality measures in place.

I think we got a little bit waylaid with hospice in that we started initially focusing on process measures. And when you look at those process measures, did someone get pain assessment? Did someone get a dyspnea assessment? Very important process of care, but they topped out really quickly and they’re not providing actionable information about quality.

So, I think in the next five to seven years we’ve got a lot of hard work to go. But what’s really exciting is we have to work force to do it. We have the rising very smart investigators who are really good scientists, who are going to be able to do this work.

Eric: Let me also ask you about the new Hospice Compare website. Where you can actually look at a bunch of hospices, see how they’re doing, kind of like Nursing Home Compare. I gotta say, I look at the data and I have trouble differentiating based on that data. Like, what’s a really great hospice? What isn’t? Your thoughts on that?

Joan: Well, I think I’ve already been pretty clear on a blog in health affairs that I think the process measures on the hospice items that are topped out. And that probably, we need some new measures and we need new work on quality. My sense is looking at the data from the CAHPS Hospice Survey. It’s doing a good job of sorting hospices out. So when we look at processes of care that we would expect to be related to quality, we’re seeing that information and we’re seeing those associations.

The one thing you have to remember about hospice is, majority of hospices are doing a really good job. It’s really only a small percentage of hospices who are usually small, they usually have a very different tax status that are related with big concerns. Not making visits in the last two days of life, not providing adequate care to these patients, and probably eight or nine percent of the people.

I think when you take a look at the quality measures that we reported previously, in JAMA, back in 2004. We saw really striking differences among people dying in hospice services, compared to people who are dying in a nursing home or dying in acute care hospital. So there’s really some large big effect sizes here. Now, we’re just not seeing that extreme variation of hospices, so that’s why I think we really need to move to a population-based instrument that measures into life care.

And for your, readers, listeners out there, I have a two year old Labradoodle who is looking me and saying-

Alex: Welcome to the GeriPal podcast, Labradoodle.

Joan: Bean!

Eric: Wait, what’s his or her name?

Joan: His name is Bean. I’ll give you another weird fact. Since 1990 I’ve named every one of my dogs from my favorite drug, which is caffeine. So we have Java, we Mocha, we have Cappuccino, and we now have been Bean, and he’s a jumping Bean. So he’s looking at me and giving me the evil eye, saying, “Let’s go outside and play chuck it!”

Eric: I got one last question for you, Joan. With the recent change of how hospices actually get paid, kind of less than the middle more on either side. Have you seen or heard of any kind of changes on how hospices are deciding who enrolls or how they care for patients? Any untold or positive aspects of that recent change?

Joan: You know, I think that’s really the subject of some work. It’s recalled that that was implemented in 2016 and the data is only getting to the point where that they can start really looking at it. The one thing I would really ask the listeners to this podcast is, I just think the visits in the last two days of life are so critical and are strongly related to perceptions of quality care.

I’ve been doing focus groups now probably, for more than 30 years. And the one consistent thing that comes out in those focus groups is, for those persons who are dying at home, family members talk about how hard it is to be the caregiver and to bear witness to someone who they love is dying. And that those visits are just really critical to really improve and enhance the quality of care. Obviously-

Eric: Bean agrees with you.

Joan: Bean is looking at me and he’s saying, “It’s really critical that he gets outside to play with his balls.”

Alex: Bean is like, “Stop the podcast already!” All right.

Alex: All right Bean, we’ll give you your wish.

Eric: Well, Joan, I want to thank you tremendously for joining us today.

Alex: Thank you so much Joan.

Eric: That was wonderful. Great insights and really great paper. And thank you for saving a little space in those journals for us in the future.

Joan: I have no doubt that you guys will do even better job of filling the journals. And hopefully we’ll start telling more of a positive news story on how things are improving.

Alex: And I’m hoping it’s not true and that we do see more Joan Teno papers.

Eric: Well, Alex, do you want to end with a little bit more of the song?

Alex: A little bit more of the song here. [Singing]

Eric: Again, thank you Joan. Thank you Bean, and thank you to all of our listeners for joining for this GeriPal podcast. Do us a favor, if you’re listening to us on iTunes or any other service, please take a moment and rate us. Unless of course, you don’t like us then yeah, you don’t need to. But again, we look forward to talking with you next week.

Alex: Bye folks.