The Affordability Crisis Behind Value-Based Care w/ Dr. Michael Chernew, Harvard

Welcome to Healthcare Rounds, the podcast where we talk with healthcare executives, policy analysts, and investors about the business, policy, and technology forces shaping the industry. If you're working in healthcare and want unfiltered conversations like this each week, please make sure to follow the show on Spotify or Apple Podcasts so you don't miss upcoming episodes. I'm John Marchica, CEO of Darwin Research Group, and today my guest is Dr. Michael Chernobyl, professor of healthcare policy at Harvard Medical School, and really one of the country's most influential health economists. He's also chair of MedPAC, the Independent Commission that advises Congress on Medicare payment policy. Michael spent years studying healthcare spending, incentives, insurance design, and really the complicated intersection between cost access and outcomes. But for me, what makes his work especially interesting right now is that many of the assumptions underpinning American healthcare around fee-for-service, Medicare Advantage, vertical integration, and so on, unit value-based care are increasingly being challenged simultaneously. So with that, Michael, welcome to Healthcare Rounds. John, thank you so much for having me. I'm honored to get to speak with you. I will say that by the time this goes out, in fact, just recently, I am now the past chair of MedPAC. Okay. That's fine. He'll be a new chair soon enough, not yet. But in any case, um I am thrilled to talk about uh these topics. I think your intro hit the uh nail on the head. So thanks. Great. So uh so you have the battle scars at least from MedPac, right? Absolutely. Absolutely. So one of the questions I just wanted to start with this because I've been thinking a lot about this, which is when I got into this game years and years ago, even before starting this company, it was kind of the world of managed care. And one of the knocks on managed care is that the things that folks wanted to do, or maybe in theory, was that we just didn't have the technology. We didn't have the ability to measure certain things and track certain things. And so when I think about where we are today, measurement has gotten much, much better. Not perfect, but it's gotten much better. And and so one of the things, I forget the name of it. What is it called when you when you change something just by observing it? It's a very famous principle, and it just slipped my mind. Yes, the Hawthorne effect, right. So do measurement systems change what's being measured? In other words, have quality scores and these raft scores, benchmarks, and things, is it is it reshaping healthcare behavior in a way that let's say policymakers didn't anticipate? So I should just go with yes. Um, I'll try and expand upon yes, but the answer is is yes. Um, it's not simply just because it's being measured, though. There's a lot of money behind it now. So let's just take quality. Um you try to measure quality because you want to understand the quality of care that Americans are getting, you can go out and measure quality. But if you tell organizations you're gonna pay them based on their scores on various quality measures, all of a sudden they do a bunch of different things. The theory was always we would pay for value, we would get value, which was better than paying for volume. But it turns out sometimes it's easier to do better on your performance scores by doing things that don't involve performing better. So um the patients you select, the way you measure uh what was going on. There could have, you know, so how you record different types of um can uh treatments that you give, and just a whole slew of other measurement activities. There's a whole measurement complex now to help organizations do better, some of which involves actually doing better, but some involves documenting that they're actually doing better. And I think my personal view is that the vast bulk of the evidence suggests that um as we've gone down this quality journey for the past several decades, the quality of care that Americans getting is not all that much better. Um the other problem, of course, is you measure what you can measure, and so doing better in terms of quality tends to be doing better in terms of quality measures, but all the other dimensions of quality, diagnostic skill, um surgical skill, a whole slew of other types of things that aren't being measured, don't get measured. I'm doing you can your audience can tell I'm an economist by my eloquence. Um, but in any case, all of those things that aren't getting measured measured don't get as much attention, and so net net is not fair, quality is all that much better. Um, you also mentioned risk scores. Um, I I think you mentioned the risk adjustment scores, and the same thing happens there. If you want to know how sick a population is, you could go through an exercise to measure how sick they are. But if you pay health plans more for serving people that are sicker, and then you ask the health plans how sick are the people you are serving, you'll find a bunch of activities designed to make that population appear sicker. And there's an enormous amount of evidence that we can talk about if you want that suggests the health plans actually do a better job of documenting illness in part because the fee for service system is just pretty bad at documenting illness, particularly illness that isn't necessarily salient. Are you talking about say up coding or making sure that you're coding to the max around some, say an insurance? Or are you speaking of something else? And actually, there's a there's a follow-up to that, which is the first thing I was thinking of is are we too reliant on say proxy measures rather than really measuring what's going on? Um so I was I I don't like the term upcoding. Um it implies that a lot of coding is uh erroneously documenting disease that doesn't exist. I think um there may be some of that going on. I think it's also the case, though, that the system just doesn't document things very well. And both with the incentives and the new technology, organizations can do a much better job documenting disease, um, which may, in many cases, be more accurate, but that does often go under the code of coding. And so um I think, yes, I'm I'm talking about that phenomenon by which health plans that are rewarded for serving sicker people do a better job of documenting the illness of the people that they serve. I actually think this is largely a calibration problem. It's not that what they're um uh documenting is incorrect, it's just we've calibrated the payments based on a system that's not documenting very well, which means you can get rewards from documenting better. Um and uh you asked whether or not I think we could just uh rely less on proxy measures. And I guess my response to that is we use measurement for a whole slew of purposes. And sometimes we separate out the measurement process from the case for measurement. And my general view is we need to think more closely and deeply about what is our case for measurement and then tailor our measurement approaches towards that. So, for example, some people, a lot of what you're discussing has to do with how, say, the government pays health plans or pays hospitals or other providers based on, say, quality scores, or if they're bearing risk as, say, part of an accountable care organization, how they get compensated for bearing that risk. There's a lot of other folks that want to measure in order to support patient choice. You'll see this in, for example, the um uh hospital compare program, for example. There's a lot of these types of things. Um and the type of measurement you need if you want to support patient choice is dramatically different than the type of measurement. The level of granularity is different, for example, than if you're trying to measure for, say, payment or accountability. Other people want to measure so they can help hospitals or other providers improve. So it's sort of a part of the quality improvement process. I think that requires a whole different set of measurement activities. One of the challenges with all of this is depending on your use case, um, if you're measuring quality, you want to decide whether you should or shouldn't risk adjust. That's become quite controversial for a bunch of reasons, just for your audience's uh knowledge. I am uh on the side that for most use cases we should adjust these quality measures so that we're not adversely penalizing organizations that are serving populations that are harder to move to a higher performance score. And uh you don't want to disincent organizations from serving those populations that may uh be have a deleterious impact on your quality scores. Um, there's a counter argument. I'm not gonna make it now, I'm happy to. The problem with moving away from what I would call proxy measures is again, depending on your use case, many of these things we have to measure at scale. If you're running the Medicare Advanced program, you really have to measure at scale. It's very hard to think through how you would get to really better detailed measures that you really have done true underlying understanding of people's health status, et cetera. It's scale for that use case. And so I've actually been spending a lot of time thinking about quality measurement systems, and I'm happy to talk about it even in more detail, but I think the narrative around quality measurement started out with the premise um a colleague of mine, Eric Schneider, says um measure score, uh, pay. So essentially you measure something, then you score everybody, then you pay based on that. And um that is in many ways more appealing than see what people do and give them money to do it. Um because the the latter incense volume, uh, which has been a big problem, and I spent a lot of my time thinking about affordability. The problem with the uh quality paradigm is I think the evidence now suggests that um you know there's gonna be some exceptions, but in most cases it doesn't work that well. And it doesn't work that well for reasons that I think are inherent. It is hard to get organizations to fundamentally transform based on the amount of dollars that we have. The organizations will also often say things like, um, well, you're gonna pay me to do better, but you don't tell me how to do better. Um there's obviously a big debate about um the nuances of the risk adjustment, so that the standard story, which is now somewhat cliche, my patients are sicker, um, which makes it hard. And I think the bigger problem is there's just a lot of noise in what we can measure. And so all the focus on measurement becomes in many ways a very expensive distraction that leads to a lot of the behaviors that are you know unintended that I mentioned earlier, as opposed to people redoubling their effort about how to do better. In many cases, many of the measures are based on um treatment of patients with chronic disease. And I think we probably both agree that um having optimal treatment for patients with chronic disease is probably one of the most important things we can do as a healthcare system. Whether it saves us money or not, it's a separate issue. But it's certainly an important goal of the healthcare system to manage patients with chronic disease. But how well patients with chronic disease are managed depends a lot on what those patients do. And so, for example, um their behaviors depend on things like how much they have to pay out of pocket to take their blood pressure medication or get their colonoscopy or a bunch of other things. And that's often outside of the purview of the organizations being scored. And in many cases, it requires some coordination amongst a lot of organizations that don't have an easy way to coordinate. If you look, for example, at um say the uh the new ambulatory specialty model and how you're gonna have to attribute performance to some group of clinicians, and um there's a there's a subset of folks that believe we should integrate more, and that integration and care coordination will enable us to get around this fragmented problem of care coordination, which is not a crazy view. It's just um integration leads to a bunch of other problems with regards to pricing and antitrust and a whole bunch of other stuff that happens. So um I I think we need to really scale back and revisit many, many, many of our quality programs. The top two that I dislike, or maybe the bottom two, I don't know. It's my top from the as an economist, it's okay if I can't speak, but it's not so good if I don't know my top from the bottom. But in any case, the my the two the two quality programs that I am least enthused about are the MA Stars program and the MIPS program. Um that doesn't mean I'm a fan of a lot of the others, but those are the two that I'm least enthralled with. Well, you you've given me about 50 follow-up questions, but I want one thing I I did want to zero in on, which is common complaint that I hear when I'm doing physician interviews, is, and you touched on this, the role of the patient. What are they actually doing? And they say things like, you know, I'm on the hook. This is a perfect ACO example, right? I'm on the hook for these quality measures, and you know, I can't make my patients eat better. I can't make them exercise, or I can't make them, you know, I'm doing everything I can to affect their HBA1C scores, but uh but there's nothing that I can do. And it's kind of comes back to, and I remember many comments around patient accountability. You say, accountable care organization. Well, where is the patient in all of this? So as an economist, you know, how do you think about how those incentives work with patients as well as with the providers? And then the second part of actually I'm gonna stick with just that on the patient side because I wanted to get to the provider side as well. So I uh I as I as my answer before alluded to, I 100% agree with that sentiment. In fact, I think we too often medicalize things that are probably best dealt with in terms of public health. But um that aside, the um we're probably gonna talk a lot about what I call value-based payment, uh, which is sort of the rubric for a bunch of payment models to providers. I got started on the sort of value movement to the extent that there is one in the uh in the space of value-based insurance design, which basically argued how do you get patients to take their uh blood pressure medication or manage their diabetes? I'm not even gonna get to eat well because no one knows how to do that. I don't even eat well enough, and I'm half the time on these podcasts. But in any case, um I think the first answer is I don't know how to get them to do all that much better, but I know how to get them to do worse, which is charge them. So the whole idea behind value-based insurance design was to remove the financial barriers to better management of your chronic conditions, make your blood pressure medication, your diabetes medication, et cetera, low cost. That was the basic idea behind value-based insurance design. That was the era of Lipitor and all these things that are now dirt generic. But when I got started, your audience can't see me, you can. I'm old. And when I started, uh statins were very expensive and people wouldn't take them. Now they're cheap, it's not such a big deal. But um, I think the system does have some role in how to do this. I think there's some measurement, there's some communication stuff, but I actually am quite sympathetic with the view that the patients have a big role to play, and there's only so much time. You know, physicians are a scarce resource, even the nurse practitioners are a scarce resource. The delivery system has a bunch of scarce resources. And so there's only so much we are able to spend in the healthcare delivery sector trying to get people to behave differently. And so one can debate whether or not wellness programs work. I think the evidence suggests they don't. Um, but in any case, having the healthcare system distracted by trying to do things that you end up getting into situations like, if you got a good blood pressure test, I'm not going to measure your blood pressure again because I'm doing well. If you got a bad blood pressure test, I want to measure it more. You have to measure it before Jane, uh, you know, making some of these up, but you have to measure it before January 1st, because otherwise it doesn't count. You need all these two met, you know, there's all this craziness in the measure definition that goes into the um process of um uh creating these quality measures that requires expensive people doing a lot of work with, I think the evidence would just minimal impact um on quality. And I think, as I said earlier, the solution is not simply better measures. The solution is really rethinking what we're trying to accomplish, being a little more um humble in what we actually can accomplish, and then creating a system intended to um accomplish sort of the feasible goals. So what I've sort of thought we should do is job one, make sure no one is getting really, really bad care. Like that's a different exercise than getting someone from a four-star to a five-star MA plan. Um but anyway. So, and I I wrote down because as I was preparing for this, I was thinking about an interview that I did several years ago with um Christopher Robertson, who's a lawyer, I believe, um, but kind of operates in this space. And I think uh my memory of that interview was just railing against copies in any form or another. Now, I know you've done research in this era. In fact, you talked about it with the Lipitur in that example. Can you envision a tough economist question? Can you can you envision a world without copies? I mean, really just eliminating the barriers, that I should say, that form of barrier for patients. Well, so I can envision one because you just have to go to the UK, right? So there are worlds that they don't use cost sharing. Uh so um it's a little rainy, their accent's a little different, but I can certainly envision it. In the US, I can't. I don't see in the US, because in order to make that world work, you need to replace the cost-depressing, the the cost-retarding effect of co-pays with some other set of structures. And I don't see us now in the US moving to a world where we build those other structures in. And so uh we could debate the merits of the UK system. I think people used to love the NHS and then COVID hit, and now you don't hear as much about it. But in any case, I think there is something to be said for the dynamism and reliance on you know uh uh individual choice and markets and innovation that you see in the American system. But if you're gonna try and rely on markets, it turns out you need to have prices. And the price that matters is the price people pay out of pocket. Uh, as value-based insurance design tried to get to, um, that doesn't mean you need to pay a lot for your insulin. Right. So I think the question is can you design knees in a smarter, intelligent way? Part of the problem is now we're so fragmented that if you get a knee surgery, you say, Well, what do I pay for a knee surgery? Well, it turns out what you pay for a knee surgery depends on what you pay the surgeon, what you pay the hospital, what you pay the rehab, what you you know, everything because the the copay system is all designed to work on through these different fee schedules, which are all fragmented. So as you're going through care in what you hope is a seamless way but often isn't, but as you're going through that process, you're also moving across fee schedules, moving across organizations. That's why you hear people who want not only coordinated care, but they want organizations that are well intended to do that. And if you do have that type of integration, then you run into other problems related, as I mentioned before, competition and such like that. And the question on the table, maybe some of your audience will say, is is there a sense in which we can use technology to provide some level of virtual coordination that doesn't rely so heavily on the type of um processes we used in the past that were incredibly cumbersome? Um Yeah, well, we're I'm gonna interrupt you there because we are gonna get to technology, but I I want to stay with this notion of measurement and what we're trying, you know, maybe overmeasurement. And I'm I'm I'm wondering, does do economists or policy analysts sometimes overestimate how governable that healthcare actually is Again, the the short answer is yes, in so many different ways. It's hard in a uh in a um world of economists that are working off of sort of abstract models or worlds of policymakers that are trying to do things that say a national scale to really understand the complicated ecosystem and importantly to understand the ecosystem is going to try and reverse engineer whatever you do to work uh to their advantage. And so if you look, I won't speak at great length, but if you look at the um the saga of surprise billing, um both the problem that arose and the solution that was imposed, and the fiasco we have, which seems to universally be considered not that much better, potentially even worse than what we had before. And um I think it's because. Because it is very hard from a policy process to understand exactly how the delivery system and the people will respond in it. And it's particularly challenging when it's not as if there's one simple organization developing the solution, but you have a complicated political ecosystem negotiating out what the solutions might be in a world that tends to be very skeptical of a bunch of regulation, and part because the regulation, and I think the surprise-building example might kind of prove this, is often not that well designed. And so we do, I think, overestimate our actually, I'm going to say this the opposite way. I think we underestimate both the challenges of the implementation and the response to regulations, and we underestimate the ability of a very savvy um uh healthcare ecosystem to respond in ways that will enable them to make the most of the regulatory environment they find themselves in. Michael, over the course of your career, and I know I'm thinking about MedPack specifically and your role on MedPack, do you feel do you ever feel like you're in this cat and mouse game where you know you're you're you're you're you're always trying to get one step ahead of the ecosystem the way that you describe it? So my my joke on this, and I'm not sure how this is so the the the often metaphor for the phenomenon you're describing is um we're playing a game of whack-a-mole. Um so I I I think that captures what you're saying. My my response is yes, we need to put a plot a piece of plywood over the entire friggin' game. And what I think, if we were going to have a longer conversation, and this is now more abstract than concrete, is one of the key things that most other countries have is they rely just more heavily on budgets and managing budgets. And in the US, we tend not to do that. And that means that the moles can pop up from other little places. And the problem, I think the move towards accountable care, in fact, you mentioned that you're gonna have Elliot Fisher on. I think the move to accountable care was at its core intended to try and find an entity, say a hospital system in some of Elliott's original work that would then be accountable for clinical and economic outcomes for a population of people and not worry about all the different ways in which the fiscal pressures could kind of ooze out. Um and we kind of have that in many ways, for example, with health plans. Um it's just the way we pay them is a little challenging. But at least there's this sort of question of how you set the budgets. And it's, you know, I have tended to believe that population-based budgets have worked better than provider-based budgets. I wouldn't say that um uh, you know, I would bet my life on that statement. If you look at uh Maryland, for example, you have provider-based budgets, and there's a lot of things that move outside of the hospital, outside of where that budget is. Is the problem you need a population-based budget, or is the problem you need to bring more things into the budget? I don't know. That's your your um your audience can write in to say. The problem with population-based budgets is if you were, for example, to give the accountability to primary care groups, as happens in in many ACOs, they struggle with how to deal with the um uh behavior of specialists or hospitals that are outside of their purview. And so, what's a poor primary care person to do if every one of their patients that ends up in the hospital or somewhere is getting a lot of skin substitutes, or an oncologist they refer to as using drugs that may or may not be high value? Um and so I think there is some tension in how this works. I do think if you happen to have a big system, as in our consolidated world you have, I do think you could blend the population and provider-oriented budgeting in such a way to make it work because a lot of the big systems are indeed providing care soup to nuts, and you should hold them accountable for managing in ways that provide efficient care, which often, by the way, is not the way that organizations are managed because the uh value-based portion of their business is small, their business model historically has not been about that, their IT systems don't necessarily support that, it's a little bit challenging culturally, and so uh I think a lot of transformation would be needed. But if you could indeed uh move to a population-based budget for people who would be attributed to a health system, which was originally Elliot's idea, um, you could then perhaps move to a world where it would demand efficiency with reasonable outcomes or at least not horrible outcomes. But you know, that's not gonna happen at national scale for a long, long time. And if it did, you've got to figure out how to deal with the pricing problems in the commercial sector for these big integrated organizations. Well, I was just gonna get to that. And before we get to Medicare Advantage, which I know is uh I think near and dear to you, at least your your in your expertise, I've had folks come on here and say the only way, I'm thinking of uh Robert Pearl being one, but others who've said the only way we're gonna get out of this mess is if we have uh a bunch of Kaisers, you know, five, seven Kaiser Permanentes, right? Where you do have that alignment where you can budget in the same kind of way, I think that you're talking about your population, your your captive population within that health plan is also the same as the delivery system. Everything is all wrapped in one, right? This kind of goes back to Kaiser's kind of an old managed care model. Um, whether that's practical, I I or not, I don't know. But is that what are your thoughts on that? And just saying Yeah, so I should say in full disclosure, my mentor who got me into this field decades ago was Alan Enthoven, and that's not that dissimilar from parts of his model. I think it's hard to have a world of a bunch of competing Kaisers and there's problems when people want to switch between the systems and how you manage those systems. I don't think that is uh inherently crazy. I do think there's other types of models that could work that don't necessarily require that level of integration. Um I'm doing some work in uh Tennessee and Virginia for a delivery system called Ballad. You may know them. They're um the result of a merger between two uh systems in the Tri-City area, so Johnson City, think Bristol. Um and um they're operating under a COPA, which is a certificate of public advantage, which essentially makes the delivery system, at least in that area to some extent, function a bit like a utility model. There's price regulation, for example. Um they um they do a lot of work. I I was there at their reporting session where they were talking about all their various quality programs. I actually think it's quite impressive. Um and they don't have the full Kaiser level of integration, and I'm not sure that would work in that area, at least not the idea of having multiple competing Kaisers. You know, Kaiser works great where it works, but there's a lot of other places where Kaiser is fine, but it doesn't, you know, everyone thinks of Kaiser, you think of California. Um there's other Kaisers and they do fine, but it's not like they're so much more efficient, they drive everyone else out of the delivery system. They provide a certain style of care for better or worse. They provide um, you know, there's things about the Kaiser model I think is great. And if you talk to George Howerson, he'd tell you it was spectacular. There's other things that might not be quite as good. People can choose or not in the Antovian model with their feet, but you can't have that work quite so well in a lot of parts of the country just because of scale. But I think there is some merit um to models that have that type of integration. Um and so I think you know they they can do a lot a lot of good. And I I am as uh maybe again because I was taught by Antoven, I am reasonably positive on type of integrated models. But a lot of the old ones that had all the hips of New Jersey's and New York's, they've they've struggled in a lot of places. Yeah, yeah, yeah. So I was thinking about, so I want to turn to Medicare Advantage. I was thinking about a presentation that I saw from Mark McClellan at least 10 years ago. Um, and by the way, we're trying to get him on the show. But uh he said something to the effect of you know, at the end of the day, we're all going to be Medicare Advantage. We're all going to be, this is where Medicare is going. We're all going to be in kind of this model and paraphrasing, but that's kind of where he was going. I still have the chart somewhere in an old presentation. So I guess where I'm going with this is when I think about the Medicare world, it's kind of like if you're in value-based, you're kind of in the ACO world or you're in the MA world. But it feels like at some level, MA is winning, although maybe winning is not the appropriate word here. Um and I maybe a different way of thinking of it is it's almost like it's the, it's becoming an operating system. It's becoming a model that is, at least from the health system side, they're getting more interested in. So from a value-based insurance design perspective, and this is your field, where are we with Medicare Advantage today? And what do you think about not pitting you against Mark McClellan, but what do you think about that comment from really lit over 10 years ago? Well, I think there's this question about whether or not we're going to more Medicare Advantage because it's better, or if we're going to more Medicare Advantage because we pay it more. And again, if you look at the MedPap reports for a range of reasons, mostly related to how our risk adjustment system works, um, organizations get paid more in the Medicare Advantage program than would be spent on individuals in the fee-for-service system. And that enables them to get access to much more generous benefits. Uh so there's a lot of technical issues with how one pays Medicare Advantage. If you recall, I don't know, two decades ago, I'm not going to get to be three decades ago, something like that. There was a huge drop-off in the enrollment and managed care in the Medicare program because the payments shrunk. And so I uh as a matter of evidence, I believe the plans are roughly 10 to 15% more efficient. Um I believe they're paid, I don't know, 15 to 20% more than would be spent in fee for service. That entire gap, some gets shared with the government. You could debate the nuances of the numbers. But the point remains a lot of that goes to help uh fill in the gaps of what is a relatively uh porous Medicare benefit. So no out-of-pocket max, co-pays, coinsurance, and so people buy uh Medigap coverage. Um, and let or uh most people buy Medigap coverage or they join MA or otherwise get supplemental coverage. And so that um if you look at both the rate of growth in Medicare Advantage enrollment and the rate of growth of benefits, as soon as they put a new risk adjustment system, V28, that all slowed down. So uh I guess the the short version is our march to MA depends on what we do, it's not inevitable. And um we will see the extent to which that plays out in going forward. Does it does it change? You're thinking about compared to traditional Medicare. I'm staying in this world, not in the commercial side. Does it change or have the ability to change physician behavior more, change patient behavior more under this sort of you know, capitated design, or or is it kind of go back to what you said about well, we're paying them more and so they're gonna do more? Or play the plan, right? So they give better benefits, but the plans pay and engage the providers and the patients differently, and they absolutely can change plan and provider behavior. I mean provider and patient behavior. And I might add, they can change provider behavior for uh towards patients that aren't in their MA plan. There's a spillover. Um, they can drive, move to cheaper out of uh hospital care. They can there's a lot of stuff MA plans can do. Um, I think that's the great promise. And that's why I think they do practice medicine. Again, pick your number, 10-15% cheaper. You could convince me it's a little smaller, you might convince me it's a little higher. But ballpark, uh, I absolutely think the plans can change provider behavior and uh the system functioning overall. Yeah, I guess that kind of comes back to what we were talking about earlier and the ecosystems. And do organizations, do the Northwells and the advocates of the world, are the are they building infrastructure and building around these kind of various payment models? Or are they responding to them? In other words, how how much of these new models are they taking into account as they're investing in the future of those organizations? That's a question for you. I feel like I should look in the comments or something. Like they don't know the answer. I my short answer, you know by now that none of my answers are short, it's all relative, but my short answer is um these organizations are made up of very uh, you know, they're very complicated organizations with a lot of different people working in a lot of different ways, making a lot of different decisions. I don't think it's sort of always the uniform. I think all of them are trying to figure out how they can be more efficient. All of them are, I don't know those ones in particular, but almost all of them I know are bearing some type of risk. All of them are trying to figure out how to manage things more effectively. I do think that is probably more reactive than otherwise. I don't think they'd be doing it if there wasn't a lot of changes in the overall system. Um, but that doesn't mean that it's not sincere. Yeah, and I and maybe maybe a different way of thinking about this kind of goes back to measurement, is we is have we gone from thinking a lot about maybe starting with the financial um and then moving to the clinical to now where everything seems to be so centered around information, it makes me think of, I think you use the term um uh was it um quality, no quality complex or measurement complex or something along those lines. Um, where where is the center of gravity today? Well, I I do think relative to, I don't know, 20, 30 years ago, the information infrastructure is way, way, way, way, way more important, way, way, way, way, way more sophisticated. Um, you'll still find places where they're using facts, there's a lot of privacy issues, but there's no doubt that information is much more salient than it ever was. And um, as is always the case, the information infrastructure depends on the uses that organizations are gonna need. And so to the extent that that's risk adjustment or um quality measurement, there'll be information um platforms that enable you to do better on those types of things. And uh a lot of times it's dual use. Take ambient listening in the so there's a primary care problem we haven't really touched much, on, but you have ambient listening uh uh in for in physician visits. Now that's helpful in terms of productivity. It also can be helpful in terms of coding. Um we could have a debate about all of that, but that information is serving multiple purposes, and it is astounding to me if you go uh to any, I mean, you're an entrepreneur, I'm not, so I should ask you these questions. But if you go to any of the conferences now where entrepreneurs gather, how much of the energy is around information generation, transmission, analysis, and use? Um the number of companies that are I've learned of companies that are managing the the number of pipes that your information goes through is just enormous. And I will say um CMS is spending a lot of time trying to upgrade the the Medicare and the government information structures just in general, in a whole bunch of ways. Some of that's going to be patient-facing. I've heard them discuss QR codes, so you don't have to fill out the same. It is maddening to fill out the same forms multiple times for visits to your same doctor or to another doctor, some really basic things. You would think that you should just have like uh the equivalent of a bank card or a QR code. This is me, this is my coverage, these are my allergies. And and we are just so, you know, for a uh uh uh for a society in which information technology is moving so quickly and is so sophisticated, so much of the healthcare system is lagging behind for a range of reasons. And I do think that's becoming ground zero for a lot of what's happening. But as I sort of the theme of this is the information that's gonna be used, all these types of tools, they're all tools, they're not that and they're gonna be used based on the nature of the incentives we put in place. So if we want AI to save money, we better be sure we have the right regulations in place to get AI save money. If we put in a bunch of codes to use more AI, different types of AI services, it's not clear that's gonna save money. So um uh the systems that we put in place will be central to guiding the way in which all these new technologies will be used and the outcomes that stem from them. Aaron Ross Powell Well, my sense is today, I'm I'm kind of with you, that just the avenues, the virtual avenues of information, the companies that are providing different types of well, just blanket information services, we'll throw AI in there, compared to what's actually being executed, what's being implemented in the in the real world, in the day-to-day world, it dwarfs it. I mean, we're kind of at that imbalance of all of these potential available tools and then what is realistically available for providers to realistically use in a day-to-day basis. And I want to make sure, first of all, Michael, I want to have you back because uh you've raised more questions than than than uh I guess answers, which is great. But given the theme of where we are today, I just want to get your sort of professorial level, economist level, human-level assessment of where we are with this notion of value-based care. And if you were today, and if you were looking forward over the next five years, where do you think the biggest changes are going to come? Um so I think value-based payment, I'll distinguish it from value-based benefit design, is going to be central to how we address our affordability problems. And whether that movement succeeds depends on how we design those models. And I think the issue, and you feel this now with healthcare spending rising as fast as it's been rising, that you just feel the pressure building up. You know, I used to live in California, so you feel the you feel the pressure building up, and you think that there's a point in which the system might explode or collapse, or you know, we'll have a policy earthquake. And when that happens, no one knows how it will play out. It could go to a single-payer system. Um, who knows how that would end up playing out? And I think there's a race between all of those organizations you were talking about to actually deliver a healthcare system that is affordable, that provides care that Americans need, um, with policy prescriptions, which at some point will be throw it all out, let's just get rid of all of this. Uh, it's just not worth it. Everyone's distracted trying to figure out how to manage risk adjustment and excuse me, and we just don't have the time or the money to do that. And so that's where I think the tension is. So I think to those of you that are listening that are sort of in this space, the the question to ask yourself is can you and can those like you deliver a healthcare system that Americans are sufficiently happy with that they don't want to scrap? And um, I'm optimistic that that can be done, but I don't underestimate the challenges in doing it. And I'm happy to come back to talk before five years um to see how we're doing. Michael, it's been an absolute pleasure. I really appreciate your taking the time uh to spend with us today. Um, and again, would love to have you back. This has been, I mean, I I think I'm gonna be going over my notes from from this conversation for quite some time. So much appreciated. Thank you so much for letting me speak with you. And uh that'll do it for uh today's episode of Healthcare Rounds. And again, Michael, terrific conversation. Thank you so much for for joining us today. If you're in healthcare and found this conversation valuable, please follow Healthcare Rounds on Spotify or Apple Podcasts so you don't miss future episodes. And if you have a minute, leave a quick review. Uh that'll help us continue bringing on more leaders in the healthcare space like Michael. You can find all of our episodes at darwinresearch.com or where you get your podcasts. I'm John Marchica, and we'll see you next round.