Lessons from the risk trenches and why the "wait and see" window on VBC is closing faster than you think

PARTNERED WITH Happy Wednesday, Hospitalogists! I'm sharing my latest deep dive today. It's on Pearl Health, and it gets into something I think deserves a more honest treatment than it usually gets: the actual math behind whether VBC pencils for health systems in 2026. Let's get after it! Pearl Health: The VBC Math That Should Change Your Mind This past week, I had a chance to sit down with Pearl Health leadership and hear their hard-fought lessons learned over the years of working with health systems across the risk spectrum - FFS all the way to global cap.   More specifically, I was looking for answers from them for how health systems can develop better, sustainable care delivery models no matter where they are on the risk spectrum. Something more pragmatic: What actually drives performance in risk-bearing models? What are the most common (and costly) mistakes organizations make early? And what does it take, operationally and financially, to succeed in value-based care at scale across traditional Medicare and Medicare Advantage in 2026? What emerged from that conversation wasn’t a framework or a pitch deck, but rather some great lessons learned from seasoned vets who walk the walk every day operating in risk.    So in collaboration with their great team of folks, here are 6 lessons from Pearl Health’s CEO & Co-Founder Michael Kopko on enabling VBC at scale. I’d also welcome thoughts from folks working inside health systems on how you think about these topics.   Wherever you are on this journey, Pearl Health is a great partner and is worth a conversation talking to their leadership team about where your future strategic priorities lie. Read in Browser Lesson 1: Risk without data infrastructure is a bet, not a strategy The single biggest predictor of ACO performance is not care model sophistication. It is whether an organization can see its patients clearly before CMS does.   Practices that enter MSSP or ACO REACH without real-time claims visibility, HCC capture insight, and utilization tracking are effectively flying blind in a retrospective system. By the time benchmark reconciliation arrives, the interventions that could have changed the outcome are already a year in the past. While most visible in MSSP and ACO REACH, the underlying issue applies across risk-bearing models, including Medicare Advantage.   This is where many health system assumptions break down. An EHR does not provide a complete picture of the Medicare population. Risk requires a longitudinal, claims-based view that extends beyond the four walls of the organization, capturing ED visits elsewhere, unaffiliated SNF stays, and specialist spend that would otherwise remain invisible. Lesson 2: Benchmark design determines your destiny more than care delivery does In most risk-bearing models, financial outcomes are largely determined before care is ever delivered. The mechanics differ between MSSP, ACO REACH, and Medicare Advantage, but the principle holds: contract design and baseline assumptions shape the outcome as much as care delivery.   MSSP’s benchmark methodology rewards historical inefficiency. Higher-spending organizations often have more room to generate savings than already efficient ones. Entering risk without modeling how the benchmark evolves over multiple performance years is one of the most common and costly mistakes.   Organizations that appear profitable in Year 1 are often surprised by rebasing. Those that understand their glide path manage it deliberately, with a clear view of how the benchmark moves over time.   This is where health systems get caught. Entering a risk contract without modeling Year 1, Year 3, and Year 6 scenarios leads to mispriced expectations. The contract you sign today is tied to a benchmark that will not stay still. Lesson 3: Physicians act on clarity, not data The most common failure mode is not a lack of data. It is a lack of synthesis and actionability.   Most organizations already have plenty of data. The problem is that it rarely translates into a clear next step. A physician reviewing a report with hundreds of flagged patients and no prioritization logic is unlikely to change care patterns. More often, they disengage.   The organizations that improve outcomes make the next step obvious: this patient, this intervention, right now.   Risk changes what physicians need to know. It is not enough to surface insight. The infrastructure has to synthesize information and translate it into clear, immediate action inside the clinical workflow. Lesson 4: The highest-cost patients are often the hardest to manage In a typical Medicare population — and increasingly in Medicare Advantage populations — roughly 5% of patients account for more than half of total cost. Identifying them is straightforward. Managing them is not.   These patients are often the most complex and least engaged. They present with multiple chronic conditions, fragmented care histories, limited trust, and frequent transitions that generate cost without continuity.   Optimizing routine care is necessary but insufficient. The real leverage sits in the high-risk cohort, which requires clinical resources and workflows most primary care environments do not have in place on day one.   This is where health systems have an advantage, but only if they use it. Post-acute networks, transitional care programs, and complex care management capabilities can be powerful risk assets. In practice, they are often disconnected from the primary care layer where they matter most. Lesson 5: Downside risk drives alignment in ways upside never does Shared savings programs create engagement. Downside risk creates alignment.   The possibility of writing a check back to CMS sharpens focus across the organization. Decision-making accelerates. Clinical and administrative priorities align. Care delivery changes in more durable ways.   This is not an argument for taking on risk prematurely. It is an acknowledgment that the discipline created by downside exposure is part of what makes these models work.   For health systems, the question is not whether to take risk. It is how to sequence into it with enough infrastructure to avoid an early loss that undermines confidence. A bad first year in two-sided risk can set an organization back years. Lesson 6: Design entry in risk and value-based contracts so your organization wins Organizations often approach VBC by asking where the biggest financial upside is. Kopko argues the better question is: where can you create the first undeniable win?   The early objective is not maximizing margin. It is building organizational confidence. A successful first year creates physician trust, leadership alignment, and board-level conviction that the model works. A failed first year can shut down that momentum for years.   In value-based care, the psychology of adoption matters almost as much as the economics. Zooming Out: The Two Camps Problem If those lessons feel familiar, it’s because they show up consistently across the market, regardless of where an organization sits on the value-based care spectrum.   Let's be honest about the landscape. I’ve had conversations with plenty of you on this issue. Hospitalogists broadly fall into two camps when it comes to taking on risk, and most of you reading this know exactly which camp your organization is in. Camp One includes the skeptics. These are the "heads in beds" operators who view Medicare as a break-even-at-best line of business, hate Medicare Advantage villainizing it purely as a denials machine, accept that Medicaid is a losing position (except when it comes to supplemental payments), and see commercial volume as the cross-subsidizing margin saver. Their VBC strategy, to the extent they have one, is defensive: let the MA plans do their UM thing, play ball on low hanging fruit, and don't bet the farm on shared savings programs that may or may not pay out. Fee for service is predictable, simple, and we can invest simply in optimizing throughput while investing in high margin specialties. In many ways, this posture reflects the risks outlined earlier: uncertainty around benchmark trajectory, limited visibility into total cost of care, and a lack of infrastructure to confidently operationalize change. This camp is larger than the conference circuit would have you believe. Change is hard. Camp Two holds the believers. They've committed resources to population health, invested in navigation and longitudinal patient relationships, hired care management teams, built (or bought) analytics capabilities, and are actively pursuing risk-bearing arrangements or prepping themselves for involuntary bundled payment models. What has shifted materially in the last 12–18 months is the policy backdrop. Many health system leaders are still operating off mental models that are already a few years stale. The LEAD ACO model, in particular, changes the calculus for Camp One organizations in ways they may not have fully absorbed. The timeline for “we can wait and see” is compressing, and the cost of sitting on the sidelines is rising. But even the believers face a nagging economic problem that doesn't get talked about enough: the labor costs of doing VBC well — coders, care managers, disease management staff, quality teams — often chew up a huge chunk of whatever shared savings they generate. The return on investment, net of operational overhead, can be underwhelming. You guys know the drill. You run the numbers, get excited about the shared savings, then watch the margin evaporate once you staff up the infrastructure to actually capture it. Even here, the friction is predictable. Data doesn’t consistently translate into action at the point of care. High-risk patients require more intensive coordination than most systems are set up to deliver. And the cost of building that infrastructure often erodes the economic upside. Pearl Health's pitch is that they can work with both camps. And the reason they can, to their credit, starts with an economic argument rather than a moral one. No "we need to save healthcare" preamble. Just math. Margin Math Nobody Talks About Here's the part of the VBC conversation that I think gets underappreciated, and Dr. Weaver from Pearl laid it out pretty cleanly: the care that risk-bearing programs are designed to reduce — avoidable readmissions, ED visits for ambulatory-care-sensitive conditions, poorly managed chronic disease — is overwhelmingly terrible-margin care for health systems. So investing in these initiatives is a net financial win across both FFS and VBC programs. This is the part that often gets missed in early modeling. The patients and services that drive avoidable cost are frequently the same ones that perform poorly under FFS economics. In other words, the opportunity isn’t just in improving outcomes — it’s in eliminating structurally unprofitable utilization.   Let’s take a 93-year-old who missed her pneumococcal vaccine and ends up in the ICU for three weeks. That's a low-rent DRG on a Medicare fee schedule that doesn't come close to covering the cost of that stay. And the readmission within 30 days of a CHF discharge is a similar story. The frequent-flyer ED utilization from unmanaged diabetes is an observation case money loser. You're not exactly losing your marquee joint replacements and neuro cases here. And let’s not forget about the downcoding catastrophe playing out among payors and providers in FFS land either.   Traditional incumbent healthcare framing has always been that VBC asks health systems to sacrifice volume. And that's true. You are, by design, reducing utilization. But the utilization you're reducing is disproportionately the stuff that was killing your margin anyway. It's like being told you have to give up your least favorite food. Oh no, not the gas station sushi. If you can replace that lost volume with higher-acuity, better-reimbursed surgical and procedural work (which most systems of meaningful scale now can), the margin profile actually improves. Seen through that lens, several of the earlier lessons start to converge. Understanding your benchmark trajectory matters more because it reframes what “savings” actually represent. Identifying high-risk patients becomes more valuable because they disproportionately drive both cost and margin drag. And taking on downside risk changes behavior precisely because it forces this math into focus.   This is the argument that should be sitting in front of every health system CFO who's still in Camp One. It's not that VBC is charity work; it's that the alternative — continuing to absorb high-cost, low-reimbursement Medicare utilization in a FFS model while payers UM you into oblivion — is arguably the worse financial strategy. Dennis Hillen, who spent 12 years at Humana and another 5 running national P&L at Oscar Health, put it in stark terms: clinicians don't treat patients differently based on who's paying the bill. The complexity of VBC is structural, not clinical. And the structural economics, when you actually run the numbers, often favor the value-based approach.   I thought this reframe was great, because it moves the VBC conversation out of the "mission vs. margin" trap it's been stuck in for years. We're not talking about doing the right thing at the expense of your bottom line. We're talking about doing the right thing because of your bottom line. That's a pitch even the most skeptical CFO has to engage with. Pearl's Three-Bucket Framework If Pearl’s 6 lessons define where organizations struggle, and the margin math explains why, the next question is execution: how do you operationalize this in a way that actually works? The leadership team laid out a 3-part framework that I think maps cleanly to how Hospitalogists should be evaluating any VBC enablement partner. Bucket One: Get into the right arrangement. This is the actuarial and data science work of matching a health system to the optimal risk-bearing structure — whether that's an MSSP track, a LEAD ACO (yes, the new hotness), an MA optimization play, or even FFS risk programs. Pearl's positioning here is explicitly program-agnostic. They're not a LEAD shop or an MSSP shop; they're an "outcomes company" that helps you figure out which program structure generates the best economic lift given your patient mix, market dynamics, and risk tolerance. The consultative angle is genuine — they claim they'll tell a prospective client if none of their offerings are the right fit, which is either admirable discipline or good salesmanship. Probably both. Bucket Two: Drive performance. Once you're in the right arrangement, you need to actually deliver — on quality measures, utilization management, risk adjustment accuracy, the whole nine yards. Pearl has a peer-reviewed journal article coming out (timing roughly coincides with this piece) showing their client base achieved MLR performance that is 2-4% better than the Medicare benchmark in year one, with improvements compounding over time. For the data-driven CFO crowd, that's the kind of credibility asset that matters. It's one thing to claim you improve outcomes; it's another to have it published in a peer-reviewed venue. Show me the receipts, as the kids say. Pearl has them. Bucket Three: Cut the cost of doing VBC. This is where Pearl's thesis gets most interesting, and where it diverges most sharply from legacy VBC enablers. The traditional enablement model — Agilon, Privia, Aledade, Wellvana, pick your favorite — has generally involved deploying human capital. We'll bring you care managers. We'll bring you coders. We'll bring you disease management nurses. The problem, as Dr. Weaver noted, is that labor is expensive and doesn't scale elegantly. The operational costs of all that human infrastructure can consume 60-80% of the shared savings you generated in Buckets One and Two. I'll say that again for the people in the back: 60-80% of your shared savings, eaten by the cost of generating them. That's not a sustainable model. That's a treadmill. Pearl's bet is on technology-first enablement. Their platform handles patient-level prioritization — telling clinicians which patients need what interventions, when — while increasingly using agentic AI to handle the administrative workflows that have traditionally required human labor. The claim is 20-40% reduction in operational labor costs versus traditional VBC enablement approaches. If that holds up at scale, it's a meaningful competitive advantage and the kind of structural unlock that makes VBC finally pencil for systems that have been running the numbers and coming up short. The AI Angle (Without the Hype) I've been pretty vocal about separating AI signal from noise in healthcare, so I appreciate that Gabe Drapos framed Pearl's AI strategy without the usual breathless futurism. No "AI will transform everything" keynote energy. Just a practical framework that breaks into three layers. Layer one is insight generation — using AI to surface the signal from the enormous volume of data that flows through VBC arrangements. Which patients are at highest risk? Which interventions have the highest expected return? Where are the gaps in care documentation? This is table stakes at this point, but execution quality varies wildly across the market. Saying you do AI-powered analytics in 2026 is like saying your restaurant has a website. Cool, but how's the food? Layer two is agentic task allocation — the "work at top of license" promise that healthcare has been chasing for years. Pearl is deploying AI agents to handle scheduling, follow-up coordination, appointment booking, and other administrative functions that currently consume clinical staff time. The logic is sound: if you can get a patient's annual wellness visit scheduled via an AI agent rather than having a nurse spend 15 minutes on the phone, you've freed up that nurse to actually do clinical work. The trust curve here matters. Patients are more likely to trust an agent to book an appointment than to discuss a medical concern, so Pearl is wisely starting with the lower-trust-threshold workflows and building from there. Layer three is what Gabe calls data portability — the idea that AI agents communicating with each other can route around the interoperability problems that have plagued healthcare IT for decades. This is the most speculative of the three, but also the most potentially transformative. If agents can exchange structured data across systems in ways that EHR walled gardens have historically prevented, you unlock a whole new set of care coordination capabilities. Gabe's explicit prediction: it will become increasingly hard for EHRs to maintain their walls as agentic AI matures. I have to wonder if Epic would agree, but the directional argument is compelling. If agent-to-agent communication becomes the new interoperability layer, the EHR incumbents have a real strategic problem on their hands. We're seeing this play out across the industry right now as every major platform figures out where agentic AI fits into their moat — or threatens it. Why This Matters Right Now The timing of Pearl's push is not coincidental. CMS just dropped the LEAD ACO model, which represents the most aggressive push toward total-cost-of-care accountability since the original MSSP launch. Health systems across the country are evaluating whether and how to engage with LEAD, and the window for program selection is compressing.   But here's the nuance that matters: LEAD isn't right for everyone. Neither is MSSP. Neither is any single program. The health systems that will navigate this transition most effectively are the ones that take a portfolio approach to their risk-bearing strategy — mixing and matching programs based on market-specific dynamics, patient population characteristics, and organizational readiness. Think of it less like picking a lane and more like constructing an investment portfolio. You want diversification, risk-adjusted exposure, and the flexibility to rebalance as the regulatory landscape shifts.   That's the real Pearl Health value proposition as I see it. Not "we're the LEAD experts" (though they clearly can go deep there). Not "we're the MSSP optimization shop." It's that they can sit across the table from a health system CFO, run the actuarial models, and say: here's the optimal configuration of VBC arrangements for your specific situation, here's the technology to execute on it, and here's the data showing it works. For health systems that have been burned by VBC enablers who overpromised and underdelivered — and there are plenty of you out there — that consultative, data-backed, technology-first approach is refreshing.   The healthcare industry has spent two decades building the policy scaffolding for value-based care. The question was never whether VBC was the right direction. The question was always whether the enabling infrastructure could make it economically viable at scale. Pearl Health is making a credible case that the answer is yes — and that the economics are better than most CFOs think.   So here's what I'm watching: Pearl's tech-first model sounds great for large integrated systems with the data infrastructure to plug into it. But the real test is whether this works at a 150-bed community hospital in middle-of-nowhere Arkansas with one IT guy named Dale. That's where VBC enablement has historically gone to die. If Pearl can crack that nut — actually make the technology accessible and the economics work for the long tail of American hospitals — they're not just another enabler. They're the infrastructure layer. The Real Test The question isn’t whether risk-bearing models are expanding across the market. It’s whether your organization is positioned to make it work.   The lessons here show up in the math — in benchmark movement, in whether clinicians act on insights, and in whether high-risk patients are actually managed. Increasingly, they also show up in how effectively organizations use AI and automation to close the gap between insight and action, and to reduce the cost of doing VBC at scale.   So the real test is simple: if you pressure-test your current strategy against these dynamics, where does it hold up, and where does it break?   For teams working through that question, Pearl engages at that level — modeling benchmarks, testing assumptions, and applying AI and automation to make the economics work in practice. If that’s a conversation you’re having internally, you can reach them at info@pearlhealth.com. This essay is a sponsored post in partnership with Pearl Health. I write these posts for companies with products or missions I believe can provide value-adds for Hospitalogy subscribers, many of whom work with/for ACOs, FQHCs, integrated health systems, health plans, and other risk-bearing organizations that want to learn more about potential value-based care partners.   If you’re interested in a sponsored deep dive, please reach out to blake@workweek.com! Thanks for the read! Let me know what you thought by replying back to this email.   — Blake   🏥 Hospitalogy Membership I'm building a community of leaders in strategy, finance, and ops at hospitals and health systems to help us connect, learn, and grow together. Apply for free membership here. Get your brand in front of 68,000+ executives and healthcare decision-makers. Let's Partner → Workweek Media Inc.  1023 Springdale Road, STE 9E Austin, TX 78721   Want to ruin my day? Unsubscribe.