Subscribe
Share
Share
Embed
Each week, Health Affairs' Rob Lott brings you in-depth conversations with leading researchers and influencers shaping the big ideas in health policy and the health care industry.
A Health Podyssey goes beyond the pages of the health policy journal Health Affairs to tell stories behind the research and share policy implications. Learn how academics and economists frame their research questions and journey to the intersection of health, health care, and policy. Health policy nerds rejoice! This podcast is for you.
00;00;00;00 - 00;00;33;05
Alan Weil
Hello and welcome to “A Health Podyssey”. I'm your host, Alan Weil. More than 12 million people are enrolled in both Medicare and Medicaid, often referred to as dual-eligibles. According to the Kaiser Family Foundation, nearly 90% of those who are dually eligible live on an annual income of less than $20,000. More than 50% are people of color, and about 25% of dual-eligible enrollees have five or more chronic conditions.
00;00;33;25 - 00;01;03;05
Alan Weil
Now, integrated care programs are designed to coordinate the benefits and services provided by Medicare and Medicaid. Despite the importance of integration, only about 10% of newly eligible enrollees are in these sorts of programs. What do we know about integrated care for people eligible for both Medicare and Medicaid and how can we increase enrollment in integrated programs? That's the topic of today's episode of “A Health Podyssey”.
00;01;03;28 - 00;01;34;04
Alan Weil
I'm here with Jose Figueroa, assistant professor of health policy and management at the Harvard T.H. Chan School of Public Health. Dr. Figueroa and colleagues published a paper in the May 2023 issue of Health Affairs, examining enrollment trends and characteristics of dually eligible enrollees in integrated care programs. They do find significant growth in the share of people with dual eligibility enrolled in integrated programs, even though overall rates remain low and there are differences across demographic characteristics.
00;01;34;07 - 00;01;41;03
Alan Weil
We'll discuss these findings and their implications in today's episode. Dr. Figueroa, welcome to the program.
00;01;41;19 - 00;01;42;20
Jose Figueroa
Thank you for having me.
00;01;42;28 - 00;02;07;09
Alan Weil
I've been working on the topic of people enrolled in Medicare and Medicaid dual-eligibles for literally decades. It seems like a challenge of bringing these two programs together is one that we never seem quite to solve. I said just a few words in the introduction about the characteristics of people eligible for both programs. Can you say a little bit about this population?
00;02;07;09 - 00;02;11;09
Alan Weil
12 million plus people. What are their medical, what are their social needs?
00;02;11;27 - 00;02;36;16
Jose Figueroa
So as you mentioned, from a medical perspective, their needs are quite complex. These are people with major disability, people with physical and cognitive impairments, people with frailty, multimorbidity, including heart failure, end stage renal disease. Many people with serious mental illness, even as close to as one in third with major depression, schizophrenia, bipolar disorder and other related psychotic disorders.
00;02;37;09 - 00;03;00;06
Jose Figueroa
And from a social perspective, by the virtue that they qualify for Medicaid, they tend to be poor, which means they have higher issues related to the social determinants of health, which we all know matter, including issues of financial insecurity, limited access to transportation, more food and housing insecurity, lower levels of educational attainment, which means they tend to have low health literacy.
00;03;00;28 - 00;03;27;02
Jose Figueroa
So in addition, as you mentioned, more likely to be people of color, specifically black and Latino in the dual population, which means they face greater issues of structural to serious discrimination, systemic racism, which we know exist in our country and can affect a person's health and well-being. When you put all of this together, it means that we have one of the most diverse, medically and socially vulnerable populations to care for.
00;03;27;26 - 00;03;55;25
Jose Figueroa
Because of this, dual-eligibles require a substantial amount of health care services to keep them safe and to keep them at home if possible. This includes they need more primary care services, more specialty care. They get hospitalized more, they require a lot of post-acute care and skilled nursing facilities. And by far one of the biggest expenses is the need for long term care services, to support at the community and also in institutions.
00;03;56;06 - 00;04;11;03
Jose Figueroa
So if you look at duals, even though they only make about 20% of the Medicare population and about 15% of the Medicaid population, if you aggregate the cost of caring for them, they account for about one in $3 across both programs.
00;04;11;22 - 00;04;36;17
Alan Weil
So this is a very high need, vulnerable population. And I really appreciate at the outset you reminded us of how heterogeneous it is. You know, in policy, you sort of talk about dual-eligibles, a term that's not particularly patient centered in and of itself, but the range of medical conditions and frailties that can lead one to be eligible for both programs is quite large.
00;04;36;17 - 00;05;05;01
Alan Weil
And so we need to think about sub groups within the population if we're trying to figure out how to meet their needs. Now, we could spend the whole conversation just talking about the problems of integrating between Medicare and Medicaid, but we're not going to do that. When we talk about integrated care programs, so people are being served by two programs run by different levels of government with different benefit packages.
00;05;05;01 - 00;05;12;09
Alan Weil
Say a little more about why integration is what we're always seeking when we talk about this high need population.
00;05;12;24 - 00;05;37;27
Jose Figueroa
Yes. And so, as you know, when the Medicare and the Medicaid program were originally designed, they weren't designed as one program. Very different coverage benefits. And then the Medicaid, it's 50 different Medicaid programs across the state. And so what that means is that you have separate budgets and in the current system, as what you said earlier, you know, 90% of duals are in non-integrated models of care.
00;05;38;19 - 00;06;29;18
Jose Figueroa
There is no administrative infrastructure or financial incentive to coordinate care across both programs and keep costs down. And when we talk about care coordination, we generally mean higher quality, less fragmented care that should yield to better patient experience and ideally better health outcomes as well. And so what we, when we mean by an integrated program, it's often financial integration that we have all of the budgets in one pot of money, which means we have one organization, typically a private insurer that is fully responsible for covering all the Medicare and the Medicaid benefits for dual-eligibles.
00;06;30;02 - 00;06;56;13
Jose Figueroa
And they typically receive capitated contracts from both the Medicare side and the Medicaid side. And to this date, currently, we only have three types of what we call fully financially integrated models of care for duals. The first one is the oldest one. It's been around since the 1970s. It's the Program of All-Inclusive Care for the Elderly, typically referred to as the PACE program.
00;06;57;03 - 00;07;21;19
Jose Figueroa
Now, this is the sort of the adult day care program. It's typically for people who are 55 years and older. And you need to have nursing home level of care in order to qualify for the program. And the one thing to note is that sort of patients live at home and then they get brought and get a one stop shop for their care facility at a adult day care type center.
00;07;22;07 - 00;07;55;12
Jose Figueroa
The second one is the fully integrated dual-eligible special needs plans, typically often referred to as the FIDE or FIDE SNPs. These are Medicare Advantage contracts that assume full responsibility for the Medicaid dollars. Now there's different types of dual special needs plans, but only the fully integrated ones, the FIDEs, are actually responsible for the total cost. There's a larger group of these SNP plans called Coordinating-Only D-SNPs.
00;07;55;12 - 00;08;32;17
Jose Figueroa
And those are not responsible. They are federally required to provide some integrated services with Medicaid, but they're not, at the end of the day, responsible for what happens on the Medicaid side in terms of what dollar amount. And then the final one is that the most recent one, which was introduced in 2011 as part of a CMS demonstration project, which are the state specific Medicare-Medicaid plans referred to as MMPs, those experienced the greatest growth when they were first introduced back in 2011, up until now.
00;08;32;26 - 00;08;51;01
Jose Figueroa
But one thing to note is that these were just temporary demonstrations. They are due to end at the end of December 2025, and the federal government expects the MMPs to then turned into D-SNPs. And that's something that will be important to follow to figure out where those patients end.
00;08;51;16 - 00;09;10;04
Alan Weil
Well, I know it's a lot of detail, but it gives a picture of what is involved in trying to do this. We're going to spend some time on what integration looks like. But why don't we go to the findings of your paper. Tell us a little bit about what you found regarding trends in enrollment in these integrated programs.
00;09;10;12 - 00;09;33;29
Jose Figueroa
Yes. And so the goal of this paper was just to provide a comprehensive overview of where we were in 2013 when it comes to integrated care and where we are at the end of 2020. And one of the reasons we were very interested is because there was actually surprisingly very little information on there. Sure, you can refer to MedPAC report that maybe in one year it said this is the number.
00;09;34;09 - 00;10;00;12
Jose Figueroa
It didn't often give it, you know, give you any detail in terms of which integrated care plan it was and the type of person that is more or less likely to enroll in one. And so that was the main motivation of this paper. And what we found was that between 2013 and 2020, the proportion of dual-eligibles in integrated care plans increased from just 2% in 2013 to about 9.4% in 2020.
00;10;00;12 - 00;10;27;26
Jose Figueroa
It's a little higher, we think, now, which is why you hear a 10% number from the Kaiser Family Foundation. Most of this growth was related to the state level MMPs, which were nearly 0% when they were first created, to now 5.2% of all duals are in MMPs. Now the number in FIDEs more than doubled, but only about 3.6% of duals are in FIDEs.
00;10;27;26 - 00;10;55;11
Jose Figueroa
So a small amount. And then the PACE program, even though it's been around since the 1970s, experienced very little growth from just about 30,000 duals to a little less than 60,000 duals, which, you know, less than 1% of all duals are in MMPs by the end of 2020. A particular concern, though, that we found, our motivation was to figure out, you know, the extent of the growth in integrated plans.
00;10;55;24 - 00;11;24;13
Jose Figueroa
But what we're finding is that growth of dual-eligibles in non integrated conventional MA plans more than doubled and now cover about 14% of all duals in 2020. And then growth in D-SNPs, we were hoping we would see a lot of them in the fully financially integrated FIDEs, but growth in the D-SNPs were mostly concentrated in the Coordinating-Only, which again, does not,
00;11;24;17 - 00;11;48;19
Jose Figueroa
they are not fully responsible for all of the Medicaid dollars. So we think that perverse incentives to shift costs still exist in those plans. And so, again, not fully integrated care. And then the last thing I'll say is that over 50% of duals remain in traditional fee-for-service Medicare, where we know there's no integration on the fee-for-service side.
00;11;49;00 - 00;12;12;19
Jose Figueroa
And that is potentially problematic for this high risk population. So just step back. While there is reassuring growth in integrated programs, the growth has been small and we still have a long ways to go, over 90% not in integrated plans. And of particular concern is this rapid growth in the conventional plans.
00;12;13;04 - 00;12;50;16
Alan Weil
But I want to talk a little more about who's in these types of plans and maybe what some of the policy next steps are. We'll have those parts of the conversation after we take a short break. And we're back. I'm speaking with Dr. Jose Figueroa about enrollment and characteristics of people who are dually eligible for Medicare and Medicaid.
00;12;51;26 - 00;13;12;16
Alan Weil
The focus here is on enrollment in integrated care plans. And thus far, we've talked about the broad numbers and the fact that enrollment, although growing, is still a small share. But you also found some differences in the demographic characteristics of people more likely to be enrolled in integrated programs and others. Can you say a little bit more about that?
00;13;13;19 - 00;13;39;29
Jose Figueroa
Yes. So there are a few key findings when it comes to the type of people that tend to enroll in integrated plans versus not. The first finding is that the younger people, the younger duals, especially those with disability and those with mental illness, were less likely to enroll in some of these integrated plans, including the PACE program and the FIDEs. Age for one reason is quite an easy answer.
00;13;39;29 - 00;14;19;17
Jose Figueroa
For some programs like PACE, there's an age cutoff and in fact there's some policies being introduced in the Senate trying to figure out whether we should expand age eligibility of the PACE program. For example, you have to be 55 years and older to enroll into it and so that's one main issue. Another issue, especially when it comes to mental, the people with serious mental illness is that there's a lot of concern that even though FIDEs, for example, in the PACE program are supposed to be comprehensive and covering all Medicaid services, they may not be as good as, and comprehensive as, covering behavioral health services.
00;14;20;04 - 00;14;50;22
Jose Figueroa
And this is for a variety of reasons. One reason is that they have more restrictive provider networks in certain areas. We know that access to behavioral health specialists is challenging. Long waitlists, for example, in certain areas, in certain markets. And if you have a plan that is much more restrictive in the types of people you could see then maybe people with mental illness would rather be in a non integrated plan with absolutely no network restriction.
00;14;50;22 - 00;15;20;05
Jose Figueroa
The other thing is some FIDEs, depending on the area, can actually carve out some behavioral health services and not be responsible for them. And so then even though they are technically supposed to be covering everything, there might be a few issue, a few gaps that doesn't fully cover the needs of people with mental illness. So that's one important finding that we found. The other important finding which was interesting is that integrated plans were more likely to enroll black and Latino people.
00;15;21;04 - 00;15;48;12
Jose Figueroa
These trends tend to track what we've observed in the growth of Medicare Advantage in general. We've seen disproportionate enrollment of people of color. These plans tend to pop up in more populated counties, more urban areas. That's where more black and Latino people live. And we found, for example, in our study that if you're a rural beneficiary, particularly white rule beneficiaries, you were not likely to enroll in these plans because they're not offered where you're living.
00;15;49;00 - 00;16;02;19
Jose Figueroa
Now, to the extent that these plans provide higher quality of care, which I know we’ll talk to, many people think that this may be a reassuring trend as a means of potentially improving health equity for populations that tend to have worse outcomes at baseline.
00;16;03;02 - 00;16;25;12
Alan Weil
Well, so this is sort of the puzzle here. I guess there are two that I'd like us to spend our remaining time on. The first is you just mentioned quality of care. The concept of integration makes a great deal of sense, but do we really know if it improves the outcomes that matter to the enrollees in the programs?
00;16;26;04 - 00;17;10;04
Jose Figueroa
This is one of the most important questions that you would think we would know the answer forwards and backwards because of the amount of time, energy, effort being spent at the federal level and at the state level and across all these advocacy organizations to improve care and push more and more people into integrated programs. But the first thing I'll say is just because you have financial integration, which we know is important to align incentives to better coordinate care across both programs, it does not necessarily mean that it will lead to higher quality of care because, again, it's just aligning incentives in terms of who the payer is, but not necessarily restructuring the entire health care
00;17;10;04 - 00;17;41;22
Jose Figueroa
system that delivers the health care services to the dual-eligible population. But we all think it is important. We think it makes sense. The second thing to know is that data on integrated care programs is surprisingly very limited. There is not a lot of data out there showing which integrated care programs work or what the specific features of the integrated programs that might actually make care better for dual-eligibles.
00;17;43;01 - 00;18;06;02
Jose Figueroa
And so, but if I was to step back and we were to look at the data to date, and yes it is mixed, but there are, you know, and yes, it is plagued by methodological issues given that the data has largely relied on large or small observational data, and as in most things in health policy, the programs weren't implemented and evaluated in a randomized controlled fashion.
00;18;06;13 - 00;18;44;28
Jose Figueroa
So given all the caveats of the data limitations and the studies out there, there are some key highlights, I think. The first one is that if you look at the data, most of the data suggests that integrated models do reduce utilization when it comes to nursing home stays and some hospitalizations, and they may be doing so in some of the studies that looked at by increasing the use of outpatient care, more primary care, for example, or more outpatient specialty care and more services at home, which is arguably a good thing since we know patients rather be at home with their family members and not in an institution, skilled nursing facility or in a
00;18;44;28 - 00;19;13;07
Jose Figueroa
hospital. So we think that's a good thing for integrated plan. Also, if you look at a data, it does seem to suggest that certain high risk duals, specifically those, for example, with dementia, may benefit more from integrated plans and those without. Now on care coordination, there is some data to suggest that integrated models of care provide better preventive care services, so routine primary care services are maybe better in integrated care.
00;19;13;28 - 00;19;46;00
Jose Figueroa
Now, there's very little data, though, on whether they actually meaningfully improve patient outcomes, like mortality, being more likely to live if you're cared for in a longer period of time. So no data on that really. There’s very limited data, surprisingly, on patient experience and the patient experience data out there is a little bit mixed. You do see it at some state level MMPs that the convenience of being in one insurance program, for example, one insurance card as opposed to up to five in certain states is better for the patient, for the dual.
00;19;47;06 - 00;20;06;23
Jose Figueroa
And then one important thing, which is, you know, a lot of the reason why we push or we think we should push people into integrated services is does it actually save money? We don't actually know if it actually saves money because we know the costs of running these programs are really expensive. National data across these models has been largely nonexistent.
00;20;07;02 - 00;20;34;00
Jose Figueroa
It stays with the private insurers. And the good piece of news, and I'll just say it quickly, is that national data across these models exists now. It was an effort at the federal level to make Medicaid data available to researchers, available to policymakers, and also data on Medicare Advantage, that encounter data, that is available now. It has been getting better over time.
00;20;34;00 - 00;20;58;08
Jose Figueroa
Early years were plagued with issues. I'm sure you're seeing that from the editorial perspective on your end when people are submitting more and more papers using these data. But we think that given that there's more readily accessible, reliable data that should get better over time, we should be able to get, you know, figure out which models of care and specifically for which type of high risk dual works.
00;20;59;00 - 00;21;17;22
Jose Figueroa
But, you know, one thing I'll say the last thing is that what we do know, though, is that the current state of affairs for the over 90% of duals, the bifurcated system is just not working. We know that doesn't work. We know it's plagued with issues. We know it leads to fragmentation and we know it leads to higher costs.
00;21;18;13 - 00;21;32;04
Jose Figueroa
But which type of integrated model of care to move people to is not really fully clear yet. We just think we should be moving in that direction and hopefully in the coming years we'll have better information on that.
00;21;33;01 - 00;22;14;01
Alan Weil
Well, so if we take as a given that un-integrated care is bad, that our knowledge of the details of what types and forms of integration are best is limited, but that directionally we feel that we know enough to want more of these models. We've been talking about integration between these programs for decades. So as we bring our conversation to a close, my question to you is, based on what you're seeing, what do you think the policy levers are here that could yield the kind of integrated care options that actually have the highest potential of improving care and saving cost?
00;22;14;01 - 00;22;14;29
Alan Weil
If we can do that, too.
00;22;15;14 - 00;22;42;06
Jose Figueroa
There's probably a few things that we can do as a country. The first thing is to really make fully integrated models the primary model of care, like it should not be 90% in non-integrated plans and 10% only in integrated plans. And one quick way to do it is, you know, why are we allowing Coordinating-Only D-SNPs, which is the biggest MA D-SNP.
00;22;42;06 - 00;23;18;01
Jose Figueroa
I mean, is it arguably any different than a non integrated plan, like the whole point of integration is to be fully financially accountable for total cost of care. And so should CMS and states consider not allowing Coordinating-Only D-SNPs? And if you want to play in the D-SNP game, you have to be either highly integrated D-SNP, which is the HIDEs, which we haven't talked about, but they were introduced a couple of years ago and they're partially responsible for a big chunk of the Medicaid dollars, either the long term care dollars or the behavioral health dollars, but not the full pot of the money.
00;23;18;12 - 00;23;37;02
Jose Figueroa
Or a better, better yet, would be the FIDEs which would be fully responsible for both the Medicare and Medicaid side. And so that's one policy, one clear policy lever that can be established that, you know, if you want to play in the D-SNP game, you have to be responsible for total dollars of care. That's one approach.
00;23;38;08 - 00;24;02;01
Jose Figueroa
The second one is, you know, one thing is as soon as people become a dual-eligible, you know, for whatever reason, someone is either Medicaid only or Medicare only, and then they become dual-eligibles. Right now, there's no system. There's no good way of letting duals know which program makes the most sense for them. And it's up to them to figure out what to enroll in.
00;24;02;20 - 00;24;30;26
Jose Figueroa
And increasingly, I think a lot of people are interested in default enrollment where the default is, as soon as someone qualifies for dual eligibility, they're defaulted into an integrated plan that exists where they are and then, you know, and then they can decide, you know, voluntarily to switch out to something else. But right now, the state of affairs is usually when they qualify for one, they're often in an integrated care plan.
00;24;30;26 - 00;24;51;22
Jose Figueroa
And, you know, that takes a little bit of time and energy to figure out which one, should they switch. So should we do more default enrollment into integrated plans? And that is one way, for example, how MMPs grew so much, you know, when an MMPs in certain states, like, you know, when they first came to, they essentially had people enrolling into them.
00;24;51;22 - 00;25;30;01
Jose Figueroa
And then people can decide whether or not they wanted to stay in it or not. And then I'll say maybe one other example. So, you know, as I mentioned, more than 50% of duals are in traditional Medicare. And in the traditional Medicare world, there is a lot of activity in terms of the alternative payment models, for example, ACOs. CMS has signaled that ACOs, accountable care organizations, are the model of the future for fee-for-service, traditional Medicare beneficiaries because in some ways is also doing the same thing.
00;25;30;02 - 00;25;57;26
Jose Figueroa
It's making an institution, an organization responsible for total Medicare Part A and B dollars. And so a question is, should we be thinking about ACOs taking on Medicaid risk? I mean, they're already taking Part A and B risks, like could we push more ACOs to be responsible for the Medicaid side and create a, you know, a coordinated system from that front, not from the payer side, but from the provider side?
00;25;57;26 - 00;25;59;20
Jose Figueroa
And that's something to explore as well as.
00;25;59;21 - 00;26;22;13
Alan Weil
Well, that's great. Well, Dr. Figueroa, thank you for taking on this complex issue. And the fact that it's been around so long, I think is evidence of the challenges associated with it. But as we noted at the outset, this is a population with very significant needs. And if we can't meet their needs, we're doing something wrong. We've got two programs designed to serve them.
00;26;22;13 - 00;26;39;00
Alan Weil
We ought to be able to get those two to work together. So I appreciate your explaining the issues at hand, the contribution to the knowledge with the data that you've been able to provide and some nice policy options for us to think about. Thank you for being my guest today on “A Health Podyssey”.
00;26;39;20 - 00;26;59;24
Jose Figueroa
Thank you for having me. And I just wanted to say special thanks to my coauthors, David Velasquez is a superstar medical student at Harvard that has been working with us, and also John Orav, who's been great as well, and the Arnold Ventures Foundation for supporting our work. And thank you again, Alan, for considering our work. We really appreciate it.
00;27;02;18 - 00;27;07;08
Alan Weil
And thanks for listening. If you enjoyed today's episode, I hope you'll tell a friend about
00;27;07;11 - 00;27;08;05
Alan Weil
“A Health Podyssey”.