Why value-based care adoption remains stuck two decades in

Value-based care has been healthcare's stated destination for nearly two decades, yet the industry keeps circling back to the same question: why hasn't it arrived? Dan D'Orazio has watched that debate play out from inside the industry's strategy rooms. As CEO of Sage Growth Partners, a Baltimore-based healthcare research, strategy and growth consulting firm, D'Orazio has advised health systems, hospitals, device makers and payers on navigating shifting payment models — and he's grown skeptical that value-based care is close to becoming the dominant way the country pays for care.

For physicians, the stakes are more than academic. Fee-for-service still drives day-to-day reimbursement even inside most value-based contracts, administrative burden keeps climbing, and physicians are increasingly asked to take on financial risk for outcomes shaped heavily by factors outside the exam room. D'Orazio argues the barriers are structural — an 80-year-old fee-for-service system, disparate data, and misaligned incentives among physicians, hospitals and payers — and that neither artificial intelligence nor a resurgence of capitation will resolve them on their own.

Medical Economics spoke with D'Orazio about why value-based care adoption has moved so slowly, what's really behind renewed interest in capitation, and why he believes physician-led organizations may offer the clearest path forward. He also weighs in on where AI can genuinely help — and where it can't.

(Editor's note: The following transcript has been edited for brevity and clarity.)

Medical Economics: How would you characterize where value-based care adoption actually stands today, versus where many experts thought it would be five or 10 years ago?

Dan D'Orazio: I would go back even further and say that in 2008 and '09, when this really began, there was a lot of hope and aspiration. I don't think we're much further along than we were, and I don't believe that's because of a lack of effort. I think we have an inherently complicated, interrelated system with varying degrees — and often competing — financial and clinical models. And if you go back even further, to World War II, when fee-for-service started, and project forward to today, we're about 80 years into this healthcare system where fee-for-service has been dominant. So we're really looking at a window of 10 to 20 years within a history of 80, and that's what we're trying to make progress on. For that reason, I think the progress has been minimal, if you look at the gestalt of, you know, "everything is going to be value-based care."

Medical Economics: What are the obstacles? Why hasn't value-based care progressed the way many experts expected?

D'Orazio: There's a finance component to this, and then there's the care model, human behavior, workflows, technology, healthcare spending, physicians versus hospitals — so many different dynamics. So if we dial back a little bit: the way we finance care is still fee-for-service. Even if you look at the value-based care programs, they're still pegged to fee-for-service, and often the incentive to remain in fee-for-service, even with the penalties, far outweighs the penalty. So financially, there's not a lot of risk — or, alternatively, benefit — to the current models, comparatively speaking. Healthcare is local. Data is disparate and unclean. Technology has really only arrived in the last 20 years, and many people say the EMR has made things more difficult, not easier. So we have interoperability and data challenges — the flow of information is not easy. And then, again, there are competing incentives. If you look at models like remote monitoring — I'll go back 15 or 20 years — we had implantable cardiac devices that were capable of sending discrete data while someone slept next to a bedside monitor, and we would know if they were having atrial fibrillation asymptomatically, which is a good thing, because if you have it asymptomatically, you're going to have a stroke, and that's not a good thing. So the data from the devices was available, and the home monitors had cellular capability, but providers were only remotely monitoring probably 20% of the patients who had this implantable device — even at a time 15 or 20 years ago when Medicare was paying more for a remote visit than an inpatient visit. So we have physician behavior and workflows that were not easy for physicians. It wasn't a technology problem. We have patient behavior. We have medical-legal concerns. We have human workforce concerns. Who's going to catch that data and act on it? Is it going to go into four different portals, because there are four major device companies? Where does that data flow — who's going to intersect it? We have alert fatigue. Is the data that's coming across really believable? So when you think about something like remote monitoring, or care outside the walls of the hospital, the technology is there. It's everything around that technology — the components I just described — that make it really challenging to bring it all together.

Medical Economics: Where does this go from here? Is it possible to overcome these challenges, or will value-based care always be in the background?

D'Orazio: It's a big question. After 20 years, I think we're still treading water at its finest. So let me put this in perspective from a financial standpoint. On paper, Elon Musk just became the first trillionaire about a month ago. So here's how I think about it: how long would it take to spend a trillion dollars? He's one person. If you spent a million dollars an hour for 24 hours a day, it would take 117 years to spend a trillion dollars. To put that in perspective, healthcare spends $5 trillion every year at this point. So if you're spending a million dollars an hour, 24 hours a day, it would take about 571 years to spend that $5 trillion. It's remarkable — the country itself is about 250 years old, and it would take more than twice that long to spend one year's worth of healthcare spending.

So then the question becomes: value-based care — what are the financial corollaries between fee-for-service and value-based care? If you look at something like hospital readmission penalties, which have basically been in the 15% to 17% range, they haven't really moved much. They've gone up and down over the last 15 or 20 years. The average hospital is facing a penalty of $150,000 to $200,000 for avoidable readmissions. So — and we've heard hospitals say this — if I'm a hospital and I take that readmission, I'm still getting paid on a fee-for-service basis, and I'm not getting fully penalized for it. If I'm not at capacity, I'm still putting people in those beds and getting revenue for my operating costs. So if I'm facing $150,000 to $200,000 as a small-to-medium hospital, my average revenue could be 15 or 20 times the penalty. Even though it's a penalty, the incentive is wildly disproportionate to keep going down the fee-for-service route, and I wouldn't even consider $150,000 to $200,000 a drop in the bucket. We just talked about how we spend $5 trillion a year. So if you add up all these programs, we are so marginally making a difference.

Now, we are getting better with treating patients. There are better outcomes. We've seen better performance with bundled programs around specific conditions. I think, medically speaking, patients are very complex, and for a physician, a provider or a hospital to really be able to say "I can control or change their behavior on my own" is not realistic. When you look at social determinants of health and other factors that drive health status, only 15% or 20% is actually controlled or directed by clinical performance. The other components are unrelated to the physician's abilities, or their span of control. And then you put the consumer in this mix — we have to partake in our own health and behaviors, and that's not something we've been very successful at as Americans.

Medical Economics: If the challenges are so great, why do we keep hearing "value-based care" from payers and the government? Why keep going down this road if it isn't the panacea it was made out to be a decade ago?

D'Orazio: We spend $5 trillion a year. Healthcare costs rise 5 to 7% on average every year. The spending is out of control and unsustainable. Roughly 61% of Americans are covered by commercial insurance through their employers; the rest — that other 40% — is Medicare and Medicaid. Employers didn't get into this business to be healthcare companies, but they're bearing the cost. Every dollar we spend on healthcare competes with a dollar for infrastructure, education, paying off the national debt, right? The U.S. is the largest economy in the history of the world, and healthcare is the largest vertical in the history of the world. So when you have something growing at 5 to 7% a year, it means it's going to double every eight to nine years. If we're at $5 trillion, we're going to go to $10 trillion. We can't sustain that growth given total GDP projections and costs. So everyone says we've got to do something about this. The reality is, what we're impacting is so small in terms of total dollars that I don't think it's meaningful enough to really move the system, and because of all the other reasons I've described, it's just fundamentally so difficult.

Here's another example I often use. We often say, "Why can't we just do online scheduling in healthcare like we do for the airlines? Should be easy, right? I need a doctor — are they available?" Well, there's something called provider data, which is how accurate the data is on a provider. Health plans have this information, and they put it in a directory and say, "Joe is available, he's a cardiologist." Well, there's all kinds of different cardiology — electrophysiology, general cardiology, primary care deals with that too. So even if I wanted to go to my health insurance portal and say I want to find a cardiologist, do I actually know that I need a cardiologist? What kind? There's provider data, then there's interoperability — can I actually connect the systems — and then there's something called physician schedules, how they set up their templates and their available slots. It's not like booking a seat on a plane. I could be a very basic case, and frankly, an orthopedic doctor doesn't want to see a sprain, because they operate. So even down to something as simple as scheduling and driving capacity, there are so many intervening factors. And if you relate that to value-based care — and I'm just talking about scheduling an appointment for access, let alone how we pay for it — we have an access challenge, we have rural health deserts, and all of these things come together and make up this 80-year-old system that's wildly complex.

Back to your question of why we keep talking about it: we don't have a choice, because the spend is so tremendous. The question is how we can really make a meaningful dent. We're 10 or 15 years into this experiment within an 80-year system that is really out of control. And the other problem is that physicians have different incentives than hospitals, med device companies and insurance companies. There's not really a through line where all these parties benefit — it's more zero-sum, in terms of "I win, you lose." The average person might stay on a health plan 18 to 24 months in the commercial world, longer in Medicare Advantage. So what's my incentive to really monitor and measure that progress longitudinally — which is value-based care's real goal — and how do I benefit? It's not that physicians show up every day and say "I don't want to do the best job." That's not what I'm saying. The connective tissue in this system is real, and you can't just pull one piece out and say it's going to be fixed, because it's so intertwined.

Medical Economics: With all these challenges — you just mentioned scheduling — I hear a lot that AI is going to solve all these issues. What's your perspective on that?

D'Orazio: The first component you have to deal with is data quality in healthcare. AI will only be as strong as the data it sits on top of, and in fact, AI can make things worse, quicker, because we don't have the right data. When we think about routine tasks — call center scheduling, and voice technology is really coming up to speed — the ability to interact with an agent is powerful. AI is real, and I'm hopeful. But it does not change all the other challenges I've described. It doesn't change the incentive problem. It might make the data cleaner, if the data's foundation is clean. It doesn't change how physicians build their clinical scheduling templates. It doesn't change fee-for-service. We have a ton of administrative waste, and I think AI is powerful for removing manual tasks, but so much of this is the human dimension. What do I do as a patient that's different because of AI? Maybe I could research my condition. I might be able to schedule more easily because of call automation. I think we'll be able to do things sooner, quicker, faster. But does it get at the core of all the other issues I've described? I still don't think it changes those challenges. I think it will marginally improve the routine, way-too-high administrative burden in healthcare.

If you look at healthcare compared to other industries, and the percentage of economic impact healthcare drives, healthcare is the number one driver because it's so big. But if you look at how healthcare drives that economic growth, it's from adding people. If you look at finance, retail and other industries, you see a huge shift toward technology and automation. That's what healthcare has been missing for a long time. So we will start to see gains in automation and technology, but it won't mean that an avatar is going to treat me. I think it will help change some of the workflow, information or administrative components, but we wouldn't need GLP-1s if it were easy to change behavior. I don't see how AI is fundamentally going to change healthcare. We have a behavior challenge — these are humans, and there's behavioral science and health economics involved in all of this. So, if it were that easy, we'd all just take the GLP-1 of healthcare — take a shot and lose weight. It's not going to happen that way.

Medical Economics: I want to ask about capitation. In the '90s, HMO capitation was a huge thing, and then it didn't really work out and kind of went away. But now I'm hearing that word a lot more. Is this part of the same value-based care push? What's going on with capitation right now?

D'Orazio: You can look at the data and see incremental growth in it, and I think this administration is certainly pushing more toward data-technology mandatory programs, which makes sense on some levels. But capitation means you have all the clinical, financial and administrative capability to make it work. If you think about a Kaiser or a Geisinger, those are integrated systems between financing and care delivery, and they're closed networks — you have the ability to understand and narrow your network. That whole HMO gatekeeping process left people with a really bad taste. If you have an open network, you don't really control where people go — you have to give people choice. One of our clients, IHCS, is a fully capitated home health, DME and infusion business, and I think they impact maybe 6% of total Medicare spend in that part of the world. But if you think about all the parts that go into capitating that piece of the world, there's a ton of moving parts. You have to be able to underwrite. You have to be on the street. You have to know the data. You have to do all the coordination of care. You've got to get someone released from a hospital and get the equipment to their house. The logistics of this are daunting. And I think the other reason people think about capitation is that it puts one entity responsible. If you think about a bundled care episode, that's an episode, and you get one payment. But in capitation, the payers inherit a life — they're expected to control the clinical spend for that person. That's a really daunting task. If you look at provider-sponsored health plans — health systems that have gotten into operating or offering products in the Medicare Advantage or commercial insurance market — Becker's did an article about a year ago noting that eight of those systems have sold or folded their health plan operations. It's no easy task, and there's also a separation between clinical and financial. If someone comes into the emergency department, or comes to my office, am I really going to say I can't treat you because you have this insurance versus that insurance? So capitation, in theory, is about getting one group or entity in control of both clinical and financial outcomes. And if you look at Maryland — are you familiar with the Maryland model?

Medical Economics: No — tell me about the Maryland model.

D'Orazio: Maryland is the only state in the country right now with what we call global budgets at the hospital level. There were three or four other states that tried this and don't do it any longer, but it's coming back into vogue with the AHEAD program at CMS, where basically you say to a hospital, "Based on this data, you're going to get the equivalent of 500 admissions this year," and whether you have more admissions or fewer, you get the revenue for 500 admissions. It's sort of like a salary — it's fixed. Whether you spend too much and need a credit card, or you save enough and build a reserve, it's on you. And in theory, that's great, because you won't have to control the care yourself — you'll have these coordinated entities doing it. But if you think about all the parties I've described — in Maryland, they've gone to this global budget for hospitals, and it has certainly changed behavior. Who I interact with in my post-acute world is different. Do I have high-quality providers? Are they going to send me readmissions? Because in this world, readmissions don't mean more money — they mean more utilization of my resources. That program will spread to the physician world too, because capitation doesn't have to mean global capitation across professional services and the hospital — there are different spends. So, yes, we would love to coordinate this. CMS is the closest you can come to it, because they control the spend, and it's essentially a single payer right now, in addition to Medicare Advantage. But if you look at the commercial insurance world and employers, they don't control all these networks — they have employees in all parts of the country. So how do you capitate and control this? It's a wildly daunting task. That's the hope, to control spending, but it's really, really difficult to execute, to align the parties and the incentives, and to make sure it's not a zero-sum game, as opposed to everyone winning together.

Medical Economics: So, all this being said, where do you see healthcare in five to 10 years? Are we going to be further down the value-based care path, or will we give up on it?

D'Orazio: I think we'll see incremental gains at best. We've been at this 20 years, and I think there are areas where we're getting smarter — again, we're 10 or 15 years into this. One of the promising areas is physician-led organizations. When you look at the data from accountable care organizations, physician-led ACOs perform better than hospital-led ones. Their incentives are completely different. I think we need a rebirth of physician leadership of independent practices. Right now, about 70% of physicians are in some kind of corporate arrangement, because it's too hard to operate independently — look at rent and labor, and look at how what Medicare pays has failed to keep up with inflation over time. How could Medicare keep up with inflation? The bill's already too large. How can employers continue to spend at this rate?

If we're going to change the system, we actually have to put significant resources in place. We're still optimizing the EMR 15, 20 years later, because the fundamental data infrastructure and nomenclature weren't there to be synchronized. It's not that I don't want this to work, but if you look back 20 years and ask why we've made such little progress, you can find it in areas like specialty care. Medicare end-of-life spending is a great example — a big portion of Medicare spending comes in the last six months of life, and it was very controversial when the Affordable Care Act was being debated, with people saying "you're going to do death panels." But if you look at the amount of spending in the last six months of life compared to the quality-adjusted life it buys, the benefit is often very narrow. So why would I not want that spending if I'm not the one paying for it out of pocket — Medicare is? I don't have to make a rational decision as the person getting the bill; I just want my family member to stay alive. If I had to pay that bill myself, we don't have a market, we don't have consumerism, we don't have transparency — we're just not on the hook. Increasingly, yes, we have to pay more and more for healthcare — an employer is now paying $20,000 to $25,000 a year for a family's coverage. That's unsustainable. So, yes, "value-based care" is what we want to say — we want more value for what we spend. But whose value, who cares, and what is "the care"? I think that's the real question.

Incrementally, I think we'll get better, but we have massive other problems too, like prior authorization. It's in the news all the time, but there are about 41 to 45 different steps in a prior authorization that are highly nuanced, manual and state-led — it's not like technology can just overcome that. And you can actually be cleared through prior authorization as a physician, but not at the facility level, so someone can show up and still not get the care. All of these rules aren't going to be unwound overnight. It's not that I want to be down on this — it's just that the task of unwinding fee-for-service, and finding enough meaningful places to generate real dollar savings when we spend $5 trillion a year, is the real challenge.