The fragmentation of outpatient health care drives up the cost of care and worsens the quality of care that patients receive, posing a risk to patients’ health. On this episode of Mathematica’s On the Evidence podcast, guests James Lee of the Center for Medicare & Medicaid Innovation, Knitasha Washington of ATW Health Solutions, Bob Phillips of the Center for Professionalism and Value in Health Care, and Lori Timmins of Mathematica discuss recent research on the nature of the problem and federal initiatives that have sought to address it.
Lee is a Lieutenant Commander at the United States Public Health Service at the Centers for Medicare & Medicaid Services’ Innovation Center. He has led care delivery for two federal primary care transformation initiatives, which provided some of the data for the research about health care fragmentation that is discussed on the episode.
Washington has a doctorate in health administration and is the president and CEO of ATW Health Solutions, a management consulting company focused on improving quality, safety, and equity in health care.
Phillips is the executive director of the Center for Professionalism and Value in Health Care. He is a practicing physician in family medicine and oversees the American Board of Family Medicine's research.
Timmins is an economist and senior researcher at Mathematica who has studied health care fragmentation.
Listen to the full episode.
You get the sense that nobody really knows who you are as a whole person, right? Your grandmother gave you the one doctor as really knowing her kidneys. Another doctor maybe understands her congestive heart failure. But who actually knows her as a whole person for all her issues, not just medically but socially as well? So if providers aren't communicating or sharing the same data to managed care, a beneficiary's care will be misaligned at best or, at worst, potentially harmful.
I'm J.B. Wogan from Mathematica, and welcome back to On the Evidence. On this episode, we're going to talk about fragmented outpatient healthcare, which can drive up the cost of care and worsen the quality of care that patients receive, posing a risk to their health. We'll talk about research Mathematica conducted for the Centers for Medicare and Medicaid Services on what causes outpatient care to become so fragmented, and how some federal initiatives have sought to reduce that fragmentation. Our guests for this episode are James Lee, Knitasha Washington, Bob Phillips, and Lori Timmins.
James is a Lieutenant Commander at the United States Public Health Service at the Centers for Medicare and Medicaid Services Innovation Center. He has led care delivery for two federal primary care transformation initiatives, which provided some of the data for the research we discuss about healthcare fragmentation.
Knitasha has a doctorate in health administration and is the President and CEO of ATW Health Solutions, a management consulting company focused on improving quality, safety, and equity and healthcare. Bob is the Executive Director of the Center for Professionalism and Value in Healthcare. He is a practicing physician in family medicine and oversees research for the American Board of Family Medicine. And, finally, Lori is an economist and senior researcher at Mathematica, who has studied healthcare fragmentation. I hope you find the conversation useful.
Lori, to start us off, could you define a few terms for us for this conversation? Like, what is ambulatory care? And what do people mean when they talk about the fragmentation of ambulatory care?
Okay. So ambulatory care, maybe it's a fancy name for care that's in the outpatient setting. So things like, you know, office visits to your primary care practitioner, specialists. That would generally be considered, you know, ambulatory care as well as things like, you know, a visit to a rural health clinic or a critical access hospital, for example. Things that would not be included in ambulatory care would be things like visits to an emergency department.
Or, you know, if you have an acute admission to a hospital, if you're admitted to the hospital, that would also not be included in ambulatory care. And when we're talking about fragmentation of ambulatory care, what we're really talking about is the receipt of care from multiple different ambulatory care practitioners.
And when we say that ambulatory care is highly fragmented, it really means that no single practitioner is accounted for a substantial proportion of those ambulatory care visits. And, you know, we'll probably talk about this later on in the podcast, but a related measure is continuity of care and continuity of care that's really focused on, you know, reoccurring visits with a given individual practitioner. Usually, you know, on models, we want to increase continuity of care. So we have a high proportion of your visits with a given individual practitioner, you know, so that they can follow your trajectory of care. Fragmentation is a related measure, but it's a bit more nuanced in that it's really looking at the spread of visits across different practitioners.
So just for example, you can have two patients that have the same continuity of care. They could have 50 percent of all their ambulatory care visits with one primary care practitioner, for example. But for the remaining 50 percent of their visits, you know, Patient A may have two -- they might see two other practitioners for that remaining 50 percent. But the second patient might see, for example, four different practitioners. So, in that case, we would say that, you know, individual, the second individual who has more spread across practitioners is receiving more highly fragmented care.
And can I ask one quick follow-up -- follow up question, which is, why is that a problem? Like, what -- why might people be concerned about highly fragmented care?
Yeah. I'm sure that Dr. Phillips probably has some thoughts on this too. But the literature suggests that, you know, individuals who have highly fragmented care, there is a risk that, you know, there's more medical testing that is done that's unnecessary. There's risk that, you know, some medical information is slipping through the cracks. You know, if you have more practitioners that you have to loop in each time, that, you know, there could be some things that are missed.
And, you know, previous literature shows that it's been -- that it's been linked to higher, you know, number of acute inpatient admissions, more costly. And also, you know, even my colleagues, they have research -- or research that shows -- that indicates that, you know, individual patients prefer it when their primary care practitioner provides more, you know, comprehensive care, that not everything is farmed out to the specialists.
Okay. Excellent. Knitasha, I'm going to turn to you. Everyone listening to this podcast has been a patient at some point and probably will be again in the future. So why might patients/listeners want to see less fragmented care?
Well, I'll put it to you like this. I have been in the healthcare space for over 30 years now. And listening to Lori just break that down was enough, right? The reality of it is, is that it's a lot. It is a lot. And even for someone who has worked -- who works in a space, who understands care delivery models, right, who focuses on quality improvement, taking a step back just to really get to those very basic definitions is -- could be overwhelming. And can you imagine as a patient, right, how they might comprehend that.
The second part I would say is that patients don't see care delivery or healthcare in the way in which we describe it or we inscribe it to them, right. It's -- you know, I'm in the hospital; I'm not in the hospital. I'm going to a doctor's visit, right. I have a follow-up. And so patients see care delivery, honestly, through a singular lens. And it's not about, you know, the -- what specialist I had to go to or the -- I was in the hospital for a couple of weeks. But it's how all of that comes together with one outcome or with my health outcome, right, how all of that comes together.
And so, you know, thinking about the fact that patients don't see it like that I think is fundamentally the reason why we have a problem. And then I'll also say to one of the points that Lori made with regards to continuity of care because all of us on the line, understand the real importance of that, a continuity of care even can be warm handoffs, right, from one caregiver to another.
And, as a patient, if I don't understand that piece of it and understand that it's also important that there's a warm handoff that's done between my primary care and the first time that I go to the cardiologist or between the cardiologist and now you're going to, you know, maybe a pulmonary doctor, like, at the end of the day, that continuity of care does even matter in those situations. And all of these things are things that patients honestly are just not at the fore -- it's not at the forefront of their thinking.
Does it also put more onus on the patient to try to be the kind of the connective tissue between the different practitioners? I was -- you know, before we started recording, I was mentioning that I had a health issue recently. I started by seeing a doctor in internal medicine and then had a follow-up appointment with another doctor in internal medicine and then saw an ophthalmologist. And each time I had to be the one to say, Okay. This is what my conversation with the previous doctor was. Here were the things that they were asking about. Here were the things they prescribed. In an ideal world, would a lot of that information be transferred from practitioner to practitioner so that I wouldn't have to convey that information?
Absolutely. In an ideal world, in an ideal world going from one specialist to another, that there's high integrity in not only the information that's transferred so -- meaning that, when you said it once, it was transferred, 100 percent, right, to that next provider. And potentially your role in it is to, you know, validate that but not to go over it every single time I touch a new provider. I'm starting my questions all over again, my background all -- so you're absolutely correct.
And I will say that, you know, just to the genesis of your question, yes, it does. It puts a lot of accountability on patients and, in particular, caregivers, right. So when you're caring for someone who's chronically ill, when you're caring for someone who has, you know, frequent visits to several different specialists, it literally becomes a job in and of itself just to manage all that.
I can tell you I have an aunt that recently passed. And myself and my sisters were the premier individuals who were helping her navigate the health system. And we kept a journal, we'd have to keep a journal, you know. So going to a visit with her actually had a journal and a book. And we'd take notes about what was said, the medications, because we knew going back to the primary care doctor or going to that next specialist or when she's in the hospital, that information was so critical. And, quite frankly, everybody doesn't know that and doesn't necessarily have the ability to be able to comprehend and understand, you know, healthcare in that way. And, unfortunately, yes. That places a much heavier burden on not only patients but caregivers as well.
Well, I'm sorry for your loss, by the way. And I -- your story reminds me a bit of the movie The Big Sick. I don't know if any of you have seen that, but there's a scene where Holly, Holly Hunter and her husband who's played by Ray Romano are trying -- Ray Romano is furiously taking down notes as the doctor's speaking, and Holly Hunter is getting more and more impatient with her husband as he's asking all these follow-up questions for the spelling of a medication. Bob, I saw your head nodding a bit as we were talking about, you know, the accountability, the onus on the patient. How -- how do clinicians, healthcare clinicians think about fragmented care, fragmented ambulatory care and the need to address it?
Yeah. Look. If you don't mind, I'm going to start with continuity for just a second because, you know, there's three kinds of continuity. We were just talking about informational continuity. Can we all see the same information about the patient. The second is another kind of tier of ability and that's care coordination or managerial continuity. Are we all working to the same plan?
And then there's relational continuity. Do I take ownership, and am I willing to be accountable for your care and worthy of your trust as a patient? And that relational continuity is the one for which is the most evidence. I think Lori was talking about this. There's great evidence that it's what patients want. And I'll talk about that a bit in a moment. It's what clinicians want. It's more valuable to us to start from a relationship and caring for someone than having to unpack all they've been through in their life and their values and their concerns. And it predicts lower costs and utilization.
The best continuity compared to the worst continuity has 15 percent lower total cost of care for Medicare beneficiaries. There's no other measure in the system that has that kind of cost effect. And then it has impact on mortality. So there have been studies in South Korea, in the US for Medicare and two out of Norway in the last couple of years that high continuity improves life expectancy. So it's really potent, that relationship and accountability and knowledge of the person caring for you is really important.
I've seen it in the last month in my own practice with a 24-year-old who came in and said, Hey. I've been seen in pediatrics for my entire life. I really enjoyed having that relationship with a pediatrician. I can't find that as an adult. Will you be my doctor? And a 72-year-old new patient coming to me -- and I shouldn't be having new patients, by the way. I've -- I've got a full panel. But a 72-year-old who says, you know, I've seen four doctors in the last two years. Will you be my doctor? So there's a hunger for this at both ends of the age spectrum, not always so much at the younger but people who've experienced it, they want it again.
And fragmentation means that they have to start that story over again and again and again. And someone has to take the time and be willing to be accountable to try to figure it out the next step and what to do and, you know, what's dangerous and what's safe for you to take. So fragmentation just makes all of that so much more complex. And I'll finish with this story. I had an 85-year-old patient who had eight physicians in four health systems. And he would run that gauntlet over a three-month period and come back to me.
And then I would work with the pharm D in my practice to get his medications down from 12 to 7, again, and take him off his insulin because his diabetes have gone away when he lost weight, again. I mean, it was a repeat cycle of I know you have trust with all of these other seven clinicians who are caring for you, but I need to keep you safe or the system is going to harm you. And that was the example of both continuity and fragmentation in the same patient. And I -- really became important for me to manage him at the end of that gauntlet each time he ran it.
Interesting. I was going to ask -- you just anticipated a follow-up I was going to ask, which was, what are the potential costs? Do we -- can we put a -- either a financial number to it, or does it risk patients' health? And sounds like there are cases where fragmentation can result in worse care and potentially really undesirable health outcomes?
Yeah. I mean, in his particular situation, I knew that 12 years before he had been given a beta blocker, and it put him into a heart block. So he had a conduction delay and was sensitive to anything that did that. But he was on a calcium channel blocker for his blood pressure and had maintained it really well. But when he got discharged from the Naval Hospital, they put him on the wrong calcium channel blocker. And when you pass out in the second row at National Stadium, there's a lot of drama associated with that. And his cardiologist decided that he should have an implantable pacemaker. And I said, Hang on a minute. His son did too. Hang on a minute. Maybe it's the medication. And so having that kind of continuity and knowledge can help in certain situations keep someone from passing out at a ballgame.
James, I'd like to turn to you. From a federal health policy perspective why -- why from that perspective is care fragmentation a problem? And you -- how do you think about the stakes? And, you know, how urgent of a problem is it? Like, is this something that we need to get right, and we need to get right now?
I think I would like to start by taking it all back to I guess, the mission of the Center of Medicare, Medicaid, or CMS. And that really is to provide the best possible care that, you know, we can, given the taxpayer dollars that we have. And care fragmentation is seen as a challenge or a barrier to both of those tenets. You know, as mentioned earlier by Bob, Knitasha, and Lori, care fragmentation potentially could delay diagnosis and therapy and cause duplicative care, obviously, you know, culminating to unnecessary and avoidable burdens for both the beneficiary or the patient, caregivers, the practitioners, practice, organizations as a whole.
So it is very important to CMS that we address care fragmentation because it directly influences the quality of care, quality of life for the patients and supports the continuation of our society's core belief that the elder members of our society continue to have access to high-quality care. So, in terms of the stakes, you know, from a public health perspective, the stakes are very high. And they're the lives of our current and future beneficiaries and also the future of our elder healthcare program as a whole, right, to maintain the stability, the funding source and, you know, to make sure that future beneficiaries will also have access to this care.
From the beneficiaries' perspective, you know, you have to just simply look at any elder member of your family or friends, your grandparents, even yourself personally if you're of that age, that if you have fragmented care, you get the sense that nobody really knows who you are as a whole person, right? You know, your grandmother may view the -- as one doctor as really knowing her kidneys. Another doctor maybe understands her congestive heart failure. But who actually knows her as a whole person, right, for all her issues, not just medically but socially as well.
So if providers, you know, and specialty care, you know, all providers aren't communicating or sharing the same data to managed care, a beneficiary's care will be misaligned, at best, you know, or, at worst, potentially harmful. Providers that don't communicate well won't have the same goals of care. And the result is harm to the beneficiary and increased costs of care. Obviously, that, you know, increased cost is very important to CMS because then it puts the entire Medicare program at risk. And, again, it harms the potential access to current and future generations of beneficiaries.
Through the making the -- in terms of the sol -- the insolvency or potential insolvency of the Medicare trust fund. Okay.
Exactly. Exactly. So, you know -- and, again, you know, one of the primary goals for CMS is that, you know, the core belief is elderly should have access to high-quality care and should keep that care and that we continue to have funding to pay for that care, just like you said. So, I mean, how urgent it is, you know, you've just got to ask any of the beneficiaries, your grandparents and the elders in your life that are receiving fragmented care, it's a huge problem, right?
For, you know, CMS, we need to continually improve on care and be very vigilant on the dollars that we spend, which means addressing care -- addressing care fragmentation and preventing it as well. And, obviously, we can't, you know, address it by restricting or limiting care, right? If a patient needs to see a specialist, of course we want them to have access to that specialist. So our approach is to only improve the quality of care per dollar we spend. And so care fragmentation, you know, that was a long-winded answer, but it's a major objective for CMS. It's not the only objective, but it is a major one.
Lori, what have we learned about the causes of fragmented care? And did your research suggest that federal initiatives like CPC+ have helped?
Yeah. So that's a good question. So in terms of the causes of fragmentation, myself and my colleagues, we've looked. One thing we want to know is, you know, what causes fragmented care. And we looked at Medicare fee-for-service claims data so Medicare beneficiaries who, you know, tend to be more -- you know, higher users of healthcare. And we, you know, combine that with rich data.
We had beneficiary data, you know, types of conditions they have, age, race, other demographic information on the physicians who treated them. So things, you know, like the gender of the physician, how long they've been in practice, as well as a bunch of primary care practice characteristics so things like, you know, are they independent practices? Are they part of an organization? Are they large practices, as well as the characteristics of the communities in which these practices are? So a lot of rich data.
And, you know, there hasn't been a lot of quantitative research that has really looked at, you know, what can predict if a beneficiary receives highly fragmented care. And I think the obvious, you know, thing that people might think is that, you know, it must be the really sick patients who are receiving highly fragmented care. You know, they have high healthcare needs. You know, maybe they see multiple specialists and whatnot. So that was very much in the back of our minds about this hypothesis. Of course, there could be other, you know, factors that are playing a role.
And before I get to the punchline of this particular study, I just want to emphasize that in our -- you know, in our study, we designated -- I'm not going to go into, you know, nuanced details, but we designated about 40 percent of Medicare fee-for-service beneficiaries as receiving highly fragmented care. And these beneficiaries, in a given year, they received eight -- sorry, 13 ambulatory care visits in the year so more than once a month on average, and they saw seven different unique practitioners. So seven different, you know, doctors. That can be, you know, nurse practitioners, what -- whatnot but ambulatory care.
And out of that 13 and 7, 8 of those visits were to specialists, and they saw five different specialists. So, you know, that indicates, you know, that, you know, just in what we're seeing based on claims data, you know, the population, Medicare population, you know, sizeable proportion are receiving highly fragmented care.
Okay. That's great. So that -- yeah. So that's putting some numbers to what -- how we're defining, how -- you know, quantifying the amount of fragmented care, what fragmented care actually looks like for these beneficiaries.
Exactly. These beneficiaries, like I said, that's 40 percent, you know, of the Medicare fee-for-service population; so not small. So what we did in this particular study, you know, to quantitatively measure, you know, we used statistical analyses, but what we found is that using this really rich data sort, we couldn't actually identify key predictors of highly fragmented care.
So, you know, it is true that those that had higher risk of, you know, Medicare expenditures so those who, you know, worse, you know, had more conditions, had more healthcare needs, it is true there was a small association, and they're more likely to have highly fragmented care. But what we found, which was, you know, interesting is that over 90 percent of the variation of an individual receiving highly fragmented care, it was due to unobserved factors. So things like, you know, patient preferences or behaviors of patients, as well as things, you know, outside of the primary care around things like, you know, specialists, for example, hospitals and whatnot. And that's really what is driving highly fragmented care, you know.
So, unfortunately, I can't pinpoint say, you know, it's exactly this or whatnot. And, you know, my colleagues and I, we weren't expecting that. But, you know, it really suggests that it's not just medical care need that is driving this, which on some, you know, sense means perhaps there is hope, you know, for improving highly fragmented care, that it doesn't have to be this way.
Okay. And is there -- is there any -- are there any insights about, you know, whether these initiatives have made a dent in reducing fragmentation of care at this point?
Yeah. So that was another research study that we did as part of this work is, you know, we looked at -- Mathematica, we're the evaluators of CPC+. So we looked to see, you know, in this particular study that I'm going to talk about, although we did this for all years, we looked in the first three years with these measures of continuity of care and fragmentation of care, and we looked to see whether, you know, CPC+ reduced fragmentation of care and improved continuity of care amongst ben -- Medicare beneficiaries.
And we did this by, you know, looking at their outcomes and comparing it to a match comparison group of, you know, Medicare beneficiaries who are at practices that were not participating in CPC+. And what we've found is that, you know, that those individuals with highly fragmented care before CPC+ started, they still receive highly fragmented care. So we didn't find evidence that, you know, CPC+ made much of a dent in terms of improving continuity of care, at least, you know, in the claims data that we observed over the first three years.
And, in fact, in the later years, we also, you know, found that as well, you know, which could suggest, you know, perhaps that these primary care models may have to be, you know, directly targeted at fragmentation of care if they really want to move the needle. Or it could also suggest that, you know, maybe incentivizing specialists isn't -- you know, may be part of something that they want to take into account because, you know, as James said, CPC+ are really focused on the primary care, you know, practitioners who are only part of this story, especially, you know, when I gave you that number of number of specialist visits and primary care.
Okay. Perfect. And thank you for summarizing some of those findings. And we'll be sure to include links in the show notes to all the research that we're discussing today on this episode. Knitasha and Bob, I'd like to turn back to you. Does it surprise -- and maybe, Knitasha, maybe we can start with you. Does it surprise you that the primary care transformation initiatives like CPC+ -- and I guess James didn't mention it, but we, you know, there -- I know there's another one called CPC Classic that preceded it, that they have not reduced fragmentation and improved continuity of care?
Fundamentally, I would say no. It's not a huge surprise. So back to the point that James made earlier, when you look at the larger vision, right, of what CMS as an organization does and what CPC+ was designed to do, obviously to, you know, what Lori has spoke to, there has been some incremental change, right. There's been some positive movement. And I think that those things are all well. But I think where it doesn't catch me off guard as much about, you know, the lack of monumental change or movement is because true disruption we have not entertained in our health system.
And one of those major disruptors are the voices of patients and people and consumers. And it goes back to James's point about, you know, like, who really owns, you know, the whole person, right, the relationship with the whole person? Values, beliefs, and preferences of consumers, patients, caregivers is a real disruptor. And nobody is really measuring it on a scale to really drive real change in our healthcare system. We know it to -- you know, to be a disruptor. But we haven't really utilized it in a way to drive the level of disruption that we need in the market. The other thing is, is that there's some fundamental, you know, issues of our healthcare delivery system that we haven't quite cured yet.
And just a couple things that I wrote down in my notes in listening to Lori. So when I think about values, beliefs, and preferences in my own relationship with my primary care doctor, you know, diversity, equity, and inclusion from the standpoint of provider access, today, my primary care physician, the reason why I continue to see him is because I've been seeing him for 20-plus years. And, quite frankly, he has the most comprehensive notes on Knitasha that I can find in one setting.
And because I'm healthy, that works for me to go to him once or twice a year to do my screenings and things of that nature. But my mother and I had the conversation that, as I get older, I'm sure my children will probably want to change my provider because is he truly aligned with my values and beliefs and preferences? No, he's not. He's -- he -- I'm a number that comes through his office that he knows, you know, for a very -- he's known for a very long time. But we just to the point that -- we just don't have that whole person, you know, relationship. And that is a real driver when it comes to outcomes.
The other couple of things that I wrote down is allocation of resources, right? You know, we have all of the certificate of need states. But, you know, when I think about where we are, with what specialists are in what regions and how you have that, there's no real, you know, barometer on allocation of resources and really, in a disruptive way, meaningfully shifting that so that we have adequate resources to care for all people.
And then the other thing that I wrote down that I think fundamentally, you know, we just have not disrupted enough is quality of services being provided, you know. So we know that there are, you know, providers that are not necessarily providing at the same rate as others. We have all of the STARS reports and this report and this report. But at the end of the day, have we really disrupted the environment, causing -- to actually shift many of those providers either out or to make certain that we are adequately incentivizing and bringing forward those that truly are, right.
And so when I think about, you know, the examples that Bob gave, right, of the patients that are coming in, well, he said, I shouldn't even be accepting, you know, patients. My panel was closed. And here I got all these patients. It's because patients know, right, where those partners are. When we find them, we go to them. And the reality of them, the reality of it is, is that there just aren't enough of the Bobs in the world, right, and the -- and the providers that are really truly trying to establish that level of relationships.
I think that there's just a lot of room for disruption in healthcare period. I think that starts with the patient voice. And then I think it also means getting some things right that we just really haven't done yet. That's making certain that we have more diversity in providers, right, that share values and beliefs like I do, that we are right-sizing resources across the country and that we're incentivizing that to happen in a real meaningful way.
And then quality of care or quality of services provided, we just got to do a better job. It's unacceptable to know what some institutions and organizations are putting out and for us to allow them to still maintain be in place. And the only thing that we do is manage, you know, when there's a bankruptcy filed I think is absurd. So we've really got to start disrupting the field, and that's starting with the voices of people.
So in terms of next steps, I'm hearing disruption in variety of ways, and I'm also hearing we need to clone Bob as a way forward.
Yeah, yeah. That part. Yeah, yeah.
Bob, I think in a planning call we talked a little bit about, like, what your intuitions were at the beginning of the process. Like, did you expect even before there was a research findings about the results of CPC+? Did you expect that it would reduce fragmentation and improve continuity of care? And you're looking like -- how does -- how do you think about the results so far and how the model was -- was designed in the first place?
So -- no. Thank you. So I think that a big focus of CPC+, at least for the practices, was in enhancing their comprehensiveness. And, to James's point, that's really important. So they hired behavioralists. They brought people in as care managers. And I did hope that the care management would reduce some of the fragmentation. But there really wasn't an active arm. There wasn't a function here to enhance continuity. Specifically, there was no measure of continuity. I know that's been part of the evaluation retrospectively, but there was no pressure, positive or negative, to increase continuity in the model.
And that -- that's somewhat important for the practice, but it's even more important for the health systems that many of them sit in. So my own practice has recently put in place measures of those incentives for the number of new patients you have every month, which will shred my continuity. There is pressure to not refer to a specific specialists but to the group, and the patient will see whoever they land on.
So, in the absence of have active measurement of continuity incentives, carrots or sticks, to enhance continuity to build those relationships on the front end of primary care, especially in fee-for-service, the incentives for the health systems is just to increase throughput and triage people to high cost services without, you know, completely devoid of a relationship. So, you know, I think -- I think that's where CPC+ did not help fragmentation specifically, and where I hope new payment models will focus and really instruct those health systems that where primary care is increasingly employed to make continuity an important high-priority measure.
Okay, great. James, I want to turn back to you. As we've mentioned in the past year, so Mathematica has published a few peer-reviewed papers that provide evidence on ambulatory care fragmentation. What can -- what can CMS do with this information? What can it do to reduce fragmented care? And does it provide some ideas for next steps? Like -- or, actually, beyond just next steps. In the future, are there ways in which the agency is already using the findings to inform policy or practice?
Yeah. I think it's important to, I guess, visualize that the Innovation Center or, you know, Center of Medicare, Medicaid Innovations, which is the longer name for CMMI, you know, we're designed to take inputs from absolutely the best sources, the newest evidence literature to make informed decisions for policy and future models. So, yes. You know, these peer-reviewed papers are absolutely a source that we utilize in our decision-making.
But it's also important to remember -- and, to Knitasha's credit, do -- we always bring it back to the beneficiary, to the patient, right, because we want to hear, I guess, from their voices, you know, how is the progress of our work doing? What are the patient's priorities? What are their experiences in our models? And really represent the voice of the patient because they are not -- you know, we take it as an internal goal to make sure that their voices are heard, and we represent patients' voices.
But we always need to always do better, you know. And that is our primary focus. And -- but, you know, we also take feedback from healthcare organizations, practitioners, practices that participate in our models, you know, just to learn from them. What new strategies are -- are they testing, you know, with or without us, you know? How are these strategies implemented? What are the pain points in implementing some of our ideas or maybe, you know, some of literature's ideas, right? And how can we just overall do things better?
And, obviously, our payer partners, we don't believe that we could do this alone, you know, even though CMS is, you know, a big federal entity. But, you know, they are our payer partners, and it is really interesting to learn from them as well to see how they are approaching some of this work. So literature definitely helps inform and shape our work. But it also, you know, very much provides support for the care delivery strategies that we design and implement in some of our models. But we truly believe that the true innovators are the model participants in, you know, in the practice all around the country.
You know, they're implementing ideas that, you know, literature may not represent or, you know, ideas that we haven't even considered here at CMS. So our models in that sense, we, you know, we support those innovative care delivery -- care delivery strategies. And by, you know, giving support to, you know, financially, you know, maybe certain strategies could be best supported by a capitated payment structure versus, you know, traditionally fee-for-service, right. So something that may work in a more progressive payment model than fee-for-service. So -- and, you know, honestly, the one thing that we don't want these practices to do is continually reinvent the wheel in primary care, right.
If they've already innovated a new strategy to provide care a certain way, you know, we want to support that. You know, we want to provide evidence, whether that be from literature, from, you know, experiences from previous models, from other practices even to support that innovative care, you know, through evidence, right, and support these strategies. And this allows these practices to -- you know, to accept and understand that, yeah. This innovative strategy does work. You know, if not, we can expand it, you know. Maybe these strategies were just really implemented in a small pilot group, right.
And now, you know, with CMS support and other payer supported models, they understand that this is the way to do things. So they expand it, right, and they move on. They move on to the next problem, you know, and innovate from there as well. So, you know, and that's really truly what we want to support in our models and in our policies. You know, a lot of healthcare organizations are very conservative or, you know, hesitant to change, right. So how can we motivate that? So that was definitely one of the biggest lessons learned from CPC+, and we're trying to -- and PCF, motivate these practices in a different manner.
Okay. Perfect. Well, I did want to talk about where we go from here. And maybe we'll start with you, Knitasha. I'd love to hear from you as well, Bob. And, Knitasha, you've already kind of teased -- you've previewed I think some ideas about disruption. But what would you like to see CMS try next in terms of policies or financial incentives to reduce care fragmentation going forward?
Yeah. So great question. One, we know that health disparities cost our country over $300 billion and would love to really see a focus where we are explicitly looking to eliminate disparities and honestly reduce overall costs because that's really also part of CPC+ is to reduce costs. And I think that in what we've seen just in healthcare period, you know, we've been so focused in the aggregate that we haven't really narrowed in on some of those areas that we know that there is tremendous amount of opportunity.
And so I would say taking an approach that really strategically integrates equity and not an add-on to an already existing system. But, essentially, what I'm saying is you take the system, and you very thoughtfully break it down. And now you apply the equity lens through that entire system, as opposed to just saying, Okay. We're going to add equity on there, and we're going to add engagement on there. We're going to put a couple of thousand and ask people to do it.
Like, we can't do it that away. We really have to do this in a very meaningful way. So that's the first thing that I would say because we probably get to goal just really focusing on disparities, to be quite frank with you. That's one. And then the other I would say is, we've really got to critically look at the voices that we have at the table. So, you know, to my point about patients, one patient doesn't equal all patients. And we've really got to make certain that there's real intentionality around diversity, equity, and inclusion as it is relate to who we're getting our information from.
So I can tell you, you know, and I commonly share this about my own personal story, it's one of the reasons why I even stayed in this space is because I was only one of -- there were no other colored individuals in the room. No one else that I was talking about quality improvement that really came from backgrounds and experiences like I did. And I think that those things are important. So we have to have a very comprehensive approach to how we're doing this. And I think, you know, who we're partnering with, patients, caregivers has to include DEI in that. And it's just not good enough to say, you know, we're adding patients or the voice of patients because we've got to critically look at who we've got at the table.
So, Bob, what about you? What do you -- what would you like to see CMS try next in terms of policies or financial incentives to reduce care fragmentation?
So I think most of it we laid out in the National Academy's report on Implementing High-Quality Primary Care a couple years ago. But from the -- on the payment side, I think James has talked about a lot of them. I do think it takes a lot of time to turn the boat. So five years is probably not enough. Ten years is probably pretty important.
I do think we have to put more resources into primary care so that they don't just hire a psychologist or a psychiatric nurse but able to bring in a social worker, community health, health worker, people who can really help the patients more routinely with their care and in the community. I think it needs to come mostly as capitated so that there are pressures on the systems they are embedded in to not just reduce costs but reduce unnecessary and fragmented care.
And I think it needs to be adjusted for the community, not just for the patient's conditions but for the community's conditions. What are the social risk factors that are affecting these people's health outside of the healthcare system, and how do you get resources into the practices and into the communities like Maryland has done so that there's more resource in the communities where it's needed. That's the payment.
On the measurement side, we're pretty explicit in the report, as well, about the need to shift up more high-value measures of primary care. And I'll tell you, we've struggled not with CMMI but with CCSQ, the measure side of CMS at -- for three years of getting continuity endorsed as a measure, not just a high-value measure. So the report is pretty clear.
It doesn't name them, but it says the functions of primary care that are most aligned with what we're trying to achieve with fragmented care are continuity and comprehensiveness care coordination. We need to really emphasize those so that the health systems are given very strong signals of what is supposed to be accomplished in the primary care setting because their inclination is to go the other direction and really try and get patients into their high-cost services.
And then I think, to Knitasha's point, is getting the community voice into the process. How do we not just account for the social risks of a community but how do we get the community as a partner and figuring out what are the highest priorities, and how can the healthcare system be an agent of change in helping them solve those problems so that trust rises, so that people are more willing to come into the healthcare system when they need to and get the services they need to when they need to? So I think Knitasha is on a -- she's got a really good idea. It is the hardest part. But if we don't start with it, we're not gonna actually get there.
Okay. Alright, so, the last question I have for this group, again looking towards the future -- You know, this podcast is called On the Evidence, so I'm curious about what evidence we still need. What questions do you hope future research can answer that might inform policies or practices intended to reduce the cost of care, improve continuity of care, and improve patient satisfaction? Let's see. I believe -- Let's see. Let's start -- James, perhaps we could start with you and then Lori and then Bob?
Sure, yeah. But before I get to that, Knitasha and Bob, you know, your response to what you would like to see CMS try in the next -- in future models or policies -- I'm not going to, you know, deny if you ask me those questions, you know, is that in the next iteration of some of our models? We're always constantly working on future models, obviously, but, you know, if you ask me personally if -- and I can't really say, but I won't deny or approve anything that you say. But what I can say is you'll be, I think, very satisfied with the next coming models that we have planned.
It is tantalizing.
Watch this space.
Watch this space. Exactly. But in terms of what questions we would hope, you know, future researchers can answer is -- You know, I think Knitasha really pointed out something very important to CMS and the Health and Human Services department as a whole, is how can we really, I guess, positively impact and reduce health disparities, health equity, you know, and how to really better incorporate social determinants into, I guess, what we provide as considered traditional healthcare services in our healthcare systems, right? In terms of health equity and health disparities and social determinants, you know, it's -- Traditionally, we see an individual that's suffering from some type of illness, and, you know, we want to treat them medically or scientifically. But studies have shown that, you know, social-determinant issues has a significant impact on the health outcomes, therapeutic outcomes of an individual. So, I would very much like to see -- and I know the agency would also be interested -- is to see how we can better incorporate social determinants into healthcare. And we would definitely be interested in incorporating that into a bigger portion or a bigger role in some of our models.
So, yeah, Bob, let's turn to you now. What are you looking for from future research as it relates to, you know, continuity of care, fragmentation of care, and even improving patient satisfaction?
So, I'm interested in looking at more patient-reported outcomes measures that are related to the relationships that people have and their -- not just satisfaction with care, but their impressions about whether that care is in their best interest. I think we need more research specifically looking at -- I mean, there's -- You know, I put into the chat a couple documents about the extensive research that's been doing linking continuity to improved outcomes of all sorts. But if we embed it in a model, how does it specifically behave to change not just -- well, certainly outcomes, but the intermediate, systematic processes that build up and improve continuity? I think we need more implementation research. I mean, the evidence now -- A program that ARC ran over the last decade about coaching practices and the kinds of relationships, the kinds of instruction that's needed to help practices change faster -- to James' point, how do we get those systems to make the changes they need to the way that we would like for these new models to do?
I think we do, frankly, need some more research to help, how do we help systems move in the right direction more quickly? It is both a function and a researchable need because if we can't coach them, the incentives may not help them get there quick enough and we wind up with another null study. So, I think a lot of the research is really about that. But Knitasha's got me thinking more increasingly about how do we -- how do we study the communities, how do we study the patients? And I think we've neglected that considerably. Although community-oriented primary care is a 70-year-old model that's worked very well around the world, we've just never really implemented it in the United States. And it's researchable.
Okay. Interesting. Alright, so, we've got the beginnings of a good learning agenda coming out of today's podcast. Lori, I'll give you the last word. Well, actually, probably we'll give -- we'll give Knitasha the last word. But, Lori, let me turn to you next. What are you looking for? Like, what kind of research question excite you in this space?
Yeah, so, I think Knitasha and Bob and James made, you know, some really compelling arguments that very much align with. What I'm thinking is, you know, social determinants of health -- I think it's -- You know, researchers widely accept, acknowledge that about 80% of healthcare outcomes come from SDOH, social determinants of health.
So, I really like Knitasha's point of view of, you know, just explicitly calling that out and models, you know, take into account the holistic and how patients come from different backgrounds and, you know, that there's a lot of nuance that, you know, maybe a model isn't going to work exactly the same for everyone. And, you know, like, as someone who evaluates these programs, we do do, you know, like, what is the impact on different racial groups or different -- But I'm thinking more from a model perspective, you know, do we have to tweak it in certain circumstances, you know, for this population group and really embed it? And related to that, there's increasingly, you know, a consolidation of healthcare.
You know, there are increasingly few independent primary-care practices. They're being, you know, consumed by these larger healthcare systems with the hospital. And, you know, there are conflicting financial incentives from doing that, which I think, you know, Bob kind of, you know, hit on a little bit when he was talking about anecdotally. And, you know, we see these. As someone who evaluates, we see that there are these differential impacts, but what I'm really interested in is more like the nuance, the mechanisms -- you know, how do we get from A to -- You know, we see A, we see B, but what happens in between, you know, how we get from A to B and not just looking at B exposed or, you know, disparities exposed, but really what happens in between.
Okay. Excellent. Alright, Knitasha I'm going to give you the last word. What research questions would you like to see pursued in the future, in terms of improving continuity of care, reducing fragmentation, and getting more of that patient satisfaction, the patient-voice aspect of things?
Yeah, so, that is an awesome question, and my response would be -- is for the system to really be looking at population health, meaning, you know, the care models, the designs of how we move forward really look at populations and people with respect to my earlier comment around values, preferences, and beliefs -- right? -- and how all of that culminates into the health system that patients and families and consumers are really looking for. So, that popHealth piece, I think, is critically important. And I'd say the second dimension to that -- and this also goes back to my earlier comment -- is a real focus on eliminating disparities, right? Because when we start looking at that piece more critically, we will naturally reduce cost, save dollars, right? So, save dollars in our system. I believe it's Deloitte 2022 report shows that it's like $320 billion annually. And so we've got to crack that nut. Yeah, the disparities across our health system, $320 billion a year, right? And it's expected to go to $1 trillion a year by, I believe, 2040, right? And so what we've got to do is really critically examine that by taking these models and truly applying the equity lens across the model.
So, one of the things that has really -- that I have not been hugely support of is how models in general -- and this is not just, you know, the recent model, but just in general the way in which we take health equity and apply it as an additional, you know, strategy, tactic, statement of work that we want to address in an already-made system, right? And that we don't -- And then the other piece of it is, is that we don't require fundamentally any framework for how we anticipate or expect providers to address it. So, we allow them to say, "Oh, well, we're gonna do this," or, "We're gonna do that." And I get the reason why we approach it that way. But one of the things that I know for a fact, the reason why many providers have not addressed this issue, is because it's such a big issue with so many different tenets to it, right?
And so it's hard work. And I believe that providing a framework -- right? -- that establishes what is minimally accepted in terms of what we should be focused on and what we should be doing in this space and how the system is going to address those is important, so I think we have to have a framework and not just -- and that's the reason why I use the equity lens. We've got to take the model through the equity lens and now adjust it to really meet the needs of, you know, beginning to eliminate disparities. So, I think that those are the two most critical things that we have to do as a health system in order to improve overall care delivery.
Well, thank you so much, James, Bob, Lori, and Knitasha. Thank you so much for speaking with me today and giving us an hour of your time to talk about this important issue. Thanks again to my guests, Lieutenant Commander James Lee, Dr. Knitasha Washington, Dr. Bob Phillips, and Dr. Lori Timmins. And thank you for listening to another episode of On the Evidence, the Mathematica podcast. This episode was produced by the inimitable Rick Stoddard. In the show notes, you'll find links to some of the resources we discussed on this episode. You can catch future episodes of the show by following us on YouTube, Apple Podcasts, Spotify, or wherever you listen to podcasts. You can also visit us at www.mathematica.org/ontheevidence.
Read a press release synthesizing key takeaways from a series of peer-reviewed journal articles on fragmented outpatient care based on studies conducted by Mathematica with support from the Centers for Medicare & Medicaid Services.
Learn more about how the Innovation Center and Mathematica are advancing understanding of primary care through an evaluation of the Comprehensive Primary Care Plus model.
Learn more about an ongoing evaluation by Mathematica for the Center for Medicare & Medicaid Innovation to determine whether medical practices’ participation in the Primary Care First alternative payment model improves quality and reduces costs for Medicare fee-for-service beneficiaries.