The fundamental point of the article is no objective way to know when healthcare AI is “right.” Benchmarks measure how good a model is at taking the test, not whether that test correlates to clinical ground truth. Perhaps the most important point of the article is that there often *isn’t* a ground truth. The “right” answer often doesn’t exist.

Anyone who has actually delivered healthcare understands this point. The “right” answer is a wave function capable of existing in multiple states at once. The key is understanding all possible outcomes and their probabilities. The outcomes are highly dependent on multiple factors: the specific patient, physician, clinical situation, treatment setting, etc. They’re even dependent on factors that are arbitrary and difficult to measure such as the impact of a complication on a surgeon’s decision making (recency bias).

It’s even more complicated than that.

Subscribe now

The article highlights partial v. full knee replacements, a perfect example of medicine’s “it depends” gray areas. In the “right” patient, a partial knee can be a homerun operation. In the “wrong” patient it can be a huge disappointment that can lead a dissatisfied patient or, even worse, early revision surgery. Who’s the right patient? It’s an almost impossible question to answer on a macro level. The answer depends on the patient’s goals and expectations, radiographic findings, and surgeon experience and expertise.

And it’s even more complicated still. Some surgeons don’t believe in partial knee replacements. Why do half the job when full knee replacements work so well? The dogma regarding indications for partial knee replacement has been challenged for years. The introduction of robotics stretched indications further. Studies show slightly higher failure rates for partial knee replacements, but those studies may be skewed by older implants and worse instrumentation. Converting a partial knee to a full knee is a straightforward operation…except when it isn’t.

Partial v. full knee replacement is an argument as old as time in joint replacement surgery. Yet the right answer remains — it depends.

Physician experience and expertise is the most important and least satisfying factor determining what’s “right.” Early in your career, the tendency is to index on guidelines, evidence-based medicine, and the textbook answer you were taught in training. That’s the right instinct for any early career physician. The problem is that the more experience you gain, the more you realize how fragile all that is.

Guidelines are often vague and lack generalizability. Evidence-based medicine is based on flawed studies. Patients don’t care what’s in your textbook. That’s not an indictment of research or the medical establishment; it’s simply the truth when you’re dealing with the human condition and the difficulty of designing and conducting high-quality studies.

As you gain experience later in your career, you learn from your successes and your failures. A “perfectly” done surgery in the wrong patient can be incredibly frustrating for all involved. A guideline discordant treatment in the right patient can lead to a great outcome. The key is knowing enough to avoid the former and having enough confidence to offer the latter. Good clinicians thrive in gray areas.

That there seldom is a single clinical ground truth is the biggest reason AI won’t replace doctors any time soon. This is what we mean when we say medicine is an art, not a science. It may not be satisfying, but it’s honest. Based on recent discourse, this seems to still be the biggest gap in understanding between docs and techies. Doctors haven’t done a good job making this argument in a way that doesn’t sound defensive or protectionist. Technologists treat it as a problem that can be solved with enough data, enough compute, and a perfectly tuned model.

To be sure, while there may be no single right answer, there is often a wrong one. The art of medicine should not be used as a shield to protect low value or poor-quality care. Nor should we ignore AI’s black box problem. Neither side should be lecturing the other.

The real answer is that AI will (eventually) shrink the gap between medical art and medical science. As a tool, it will validate (or invalidate) our instincts — quantifying and crystallizing intuition while helping us navigate the gray areas. Instead of replacing physicians, it will help them do what they do best — use experience and expertise to turn noise into signal.

BS

It was an interesting year. Though I’d like to think I was intentional with the transition, I learned firsthand that career pivots don’t always go as planned. Suffice it to say, it’s a tough fundraising environment for healthcare startups. I knew going in that life in the startup world can be unpredictable, chaotic, and sometimes fickle. In fact, I factored uncertainty into my risk-benefit analysis and planned accordingly. I’m glad I did.

Despite some ups and downs, I consider the past year an overall success. Although I had to pivot from Phase 2 to Phase 2.5 on the fly, I ended up stronger, wiser, and in a better place for it. I learned the power of having a large network and am grateful to all the people who supported me along the way.

Given that it’s been a year and much has changed, I thought it might be good to provide an update on what I’m doing professionally. I heard the term “portfolio career” somewhere, and it seems apropos for where I’ve landed. Here’s the portfolio.

Subscribe now

Back in (Clinical) Action

Last week, I joined Cambridge Health Alliance, an innovative community health system in the Boston area named one of the top 10 hospitals in the nation for Avoiding Overuse and number one in Massachusetts for Health Equity, Community Benefit, and Avoiding Overuse by the Lown Institute. At the health system level, “value-based care” is oft discussed but rarely executed (unless mandated). That’s not the case at CHA (75% of its patients have some form of government-backed insurance), and I’m excited to join an organization that walks the value walk.

My practice will be non-surgical and office-based, seeing patients a couple days a week at CHA’s beautiful new office space in Medford, MA. (In all honesty, I’ve never seen a more thoughtfully designed, patient-friendly office layout in my 20+ years in healthcare). I’ve gotten to know CHA’s Ortho Department Chief, Hans van Lancker, a bit over the last few years. Hans and I share a similar entrepreneurial spirit, and I look forward to supporting his initiatives both in MSK and at the broader system level. It’s always refreshing to work with people with an innovative mindset.

Finally, CHA has a teaching affiliation with Harvard Medical School, and I hope to bring my unique experience and perspective to the next generation of physicians. After all these years, I may finally make it to Harvard! My role at CHA will be my first employed position, quite a change from my last clinical job inside a PE-backed MSO.

At the risk of being too trite, I’m optimistic my role with CHA will feel a little more “pure.” I still think having at least one foot in real world practice is important for clinicians who want to innovate in healthcare. After a year away, I’m dusting off the scrubs and white coat and heading back to the frontlines. (Ok, to be honest, I’ve still been wearing scrub pants around the house).

VBC MSK Redux

One thing that hasn’t changed over the past year — my passion for building high value, evidence-based MSK care pathways. In fact, if anything, that conviction is stronger than ever. These are interesting, unsettled times in the world of Orthopedics whether it’s struggling MSO platforms or massive reimbursement cuts.

Against that backdrop, I joined RightMove Health in June as a part-time Medical Director. RightMove is an expedited, specialist-led evaluation and care coordination platform for musculoskeletal conditions. Originally spun out of the Hospital for Specialty Surgery, RightMove’s approach is differentiated in a few ways. First, treatment protocols are backed by years of experience and expertise from HSS, widely considered to be the leader in Orthopedic specialty care.

The company partners with payers to go at-risk for MSK spend on a defined set of conditions and diagnoses. Most importantly, the company seeks to close the care loop by collaborating with high-quality surgeons rather than trying to gatekeep them. In my opinion, no one has figured out how to own the entire MSK care journey from start to finish, but RightMove is out to change that.

At RightMove, I’ll help shape hip and knee arthritis care pathways, support the clinical team of nurse practitioners and physical therapists, and chip in on strategy, outreach, thought leadership, and platform design. All the stuff I really enjoy.

I’ve known RightMove’s CMO, Adrian Thomas, for years and jumped at the chance to come on board. Though it’s only been a couple of months, I think they may be on to something!

Meetings and Thought Leadership

Never thought public speaking was my thing but turns out I actually kind of enjoy it. Over the last few years, I’ve appeared as a guest on multiple podcasts and even hosted my own in association with Commons Clinic. I’ve been a presenter, guest speaker, or moderator at a number of meetings including the AAHKS Business of Arthroplasty Course, AAOS Annual Meeting, Becker’s Healthcare Conference, Downeast Digital Conference, Duke Margolis Value-Based Care meeting, the RSNA meeting hosted by Prenuvo, DOCSF, and others. I even attended HLTH for the first time.

In the coming weeks, I’ll be presenting on the current state of MSOs at the ASSH meeting, giving two virtual Grand Rounds sessions to primary care providers at Archwell Health (in association with AristaMD), and moderating a session on AI in the ASC setting at the American College of Surgeons meeting next month. (I don’t plan on attending HLTH again this year.)

FWIW, I recently passed 39k followers on LinkedIn and continue to write The Surgeon’s Record. (My series of posts on the State of PE-Backed Ortho MSOs was especially successful). Thanks to everyone who takes the time to read, engage, and reach out. With everything else going on, I’m not sure I’ll have quite as much time to write going forward, but it’s definitely something I plan to keep doing. I’m always mindful of favoring quality over quantity in both my short form and long form writing.

No matter what, my intention is to keep challenging conventions with my typical earnest, thought-provoking, mildly contrarian approach.

Good Bones Medical Advisory

I started an LLC a few years ago when I first dipped my toes into consulting and advisory work. The work came mostly through word of mouth, networking, and writing thought pieces, and I picked up a few stipends through clinical leadership positions adjacent to my full-time medical practice.

At the time, the company was mostly a corporate entity meant to keep my 1099 income separate from my W-2 income and take advantage of a few tax breaks. For the longest time I toyed with the idea of making it something more formal but never had the time or the impetus to follow through. But given my career change and an uptick in opportunities, now seemed like a good time to make it official.

The inbound has been good, ranging from the RightMove role above, to an advisory role with equity (Mira Health), to a couple hourly consulting gigs. I also continue to advise Commons Clinic. I’m still figuring out how best to manage things on the fly, but it’s been enjoyable and engaging so far. I envision a future where GBMA becomes more of a focus and vehicle for what I’m calling career Phase 3. It’s not there yet — which is fine because I’m not ready for it to be there yet.

Committees/National Leadership

If there’s one piece of advice I would give to young docs, it’s to get involved in professional societies early in their careers. Not only does this help with network development, but it also keeps you informed and gives you a seat at the table. I continue to serve as Chair of the Practice Management Committee for AAHKS, providing resources for the Association’s membership and attending BOD meetings.

I also serve on the Practice Management Instructional Course Committee for AAOS, reviewing submissions and serving as moderator for sessions at the Annual Meeting. I continue to review submissions and serve on the Editorial Board of the Journal of Arthroplasty. The best thing about these roles is that I always learn something by doing them.

A Personal Note

On a personal level, it’s been an enjoyable year. I’ve been more present as a father — for better or worse from my teens’ point of view. Mornings are more intentional, there’s something ikigai about grinding beans and brewing a fresh pot of coffee first thing.

My body has mostly adjusted to the new schedule, and I even had Claude design a new workout routine fine-tuned to an almost 50-year-old, 6’6” former total joint surgeon with a few years of OR wear and tear on him. It’s actually worked out pretty well (pun my own). My core is getting stronger, I learned what a “Pallof Press” is, and I’ve come to begrudgingly look forward to leg day.

With us both working mostly from home, my wife and I have become officemates and coworkers of a sort. While I look forward to getting back into the office a couple days a week, I’ll miss our midday walks, late afternoon caffeine breaks, and jokes about who gets to call our home office their “Executive Suite.”

So, there you have it, my professional portfolio. Another thing I’ve learned over the past year is that there are a lot of other clinicians out there contemplating a similar career move. As always, I’m happy to connect and share my experiences. I’ve found it a challenging but rewarding path, and hopefully others can learn from it.

While my portfolio is pretty robust, there is still bandwidth for 1-2 more appropriately scoped advisory roles. I’d love to do something on the tech side or dip my toes back into practice management at the ASC or physician group level. It’s quite possible my existing roles will expand in those directions, too.

You never know which way things will go, and that’s half the fun.

BS

Last week, the governor of Massachusetts signed a bill paving the way for the state to claim eminent domain over Norwood Hospital. The current owner, Medical Properties Trust (Steward’s landlord), has until November 1st to find an “acceptable buyer” or the state is authorized to take over the bidding process.

Massachusetts has defined an “acceptable buyer” as a qualified non-profit hospital operator capable of restoring Norwood to a full-service inpatient hospital with the capital (~$150M, presumably with low leverage) to finish the clinical build out (flood damage shut Norwood down six years ago). MPT claims it’s already put $350M into the rebuild, a sticking point that’s certain to muddy the waters.

The state already divvied some of Steward’s other hospitals, committing nearly $800M of taxpayer money to rescue and rehabilitate 5 facilities. MGB, previously told by the state AG that it could no longer expand its hospital footprint, tried to buy Norwood. MPT subsequently raised the price 50% and MGB walked away. (MGB refusing to pay inflated prices? Oh, the irony!)

The state would likely prefer that a true non-profit, not a Non-Profit in Name Only (NPINO), take control of Norwood. But those systems (Boston Medical and Lowell General) are struggling to integrate the Steward Hospitals they already inherited. Expecting them to take on another distressed facility that needs extensive capital investment and physical infrastructure rebuild is probably a bridge too far.

I get the reticence to take an unconventional approach with Norwood given the damage done by Steward. But there is precedent. A few years ago, the state approved the mega-merger of Beth Israel Lahey Health. Consolidation of the hospitals was projected to increase commercial healthcare spending by $130M-230M annually. Just recently, the Health Policy Commission approved an affiliation between MGB and CVS Minute Clinic — despite a projected $40M annual cost increase. It seems at best inconsistent and at worst hypocritical to greenlight these partnerships while limiting the options for Norwood Hospital.

Here’s an idea: try something different — take everything Steward did and flip it on its head.

Real estate was one of Steward’s biggest failures. With eminent domain, the Commonwealth owns the land and can lease it back at nominal rent. Think of it as a sale-leaseback in reverse, one that removes the real estate extraction mechanism. Fix the governance issue by making physicians, nurses, physical therapists and other clinicians member-owners of a cooperative with board control, membership, and vote. Member-owners hold capital accounts instead of equity they’re looking to flip for a payday.

Outside money is still necessary, but terms can be set differently — subordinated debt, a hard cap, no votes, and no board seats. Investors get bond-like returns, not a 5–7-year exit timeline and an incentive to maximize returns. Control stays with the people delivering care; those unwilling to commit to the long-term mission and vision need not apply. Lock the asset to prevent resales or sale-leasebacks, restrict to state-approved activities, and revert the whole thing back to the Commonwealth if it fails. Public money is going into a former Steward hospital either way. Give clinicians the keys.

Finally, rethink the full-service inpatient requirement. Rather than doing to more of the same, let the operators decide what the community really needs. Commit to value-based care, radical transparency, and smart tech integration. Restore access while creating infrastructure that helps reduce demand. Traditional health systems are great at verticalization, and less great at true integration. Direct-to-employer, risk-bearing contracts, ACOs, ambulatory-first, hospital-at-home — let the hospital serve as a living experiment in how healthcare can be done right.

Norwood is in the suburbs of Boston, a region that’s supposed to be the bellwether of American healthcare innovation. We’re great at developing novel therapeutics. We have some of world’s premier biomedical research institutions. We’ve proven we can grow health systems — and raise costs in the process. What we haven’t yet proven is that we can innovate what’s arguably America’s biggest healthcare challenge: efficient, cost-effective, engaging, sustainable care delivery.

It might fail spectacularly. I'd argue that's still better than the "safe" route of capitulating to NPINOs or further overburdening safety net systems that are already stretched. Mark Cuban has argued that a hospital should be run like a startup.

Here’s the perfect opportunity.

Subscribe now

BS

A recent Relentless Health Value podcast called physicians and self-insured employers the “Sleeping Giants” of healthcare. The (excellent as always) episode features Dr. Suhas Gondi who, along with lead author Dr. Zirui Song, wrote a recent NEJM piece arguing that most clinicians are oblivious to the mechanics of self-insured employers and how it relates to the patients (i.e., employees) they treat. The “sleeping giant” label refers to a fully functioning direct contracting marketplace with two parties (employers and clinicians) taking full advantage of their latent leverage.

I’ve experienced direct contracting both as a “Center of Excellence” surgeon and as a physician executive building and executing such arrangements. I absolutely believe the process works and has enormous potential to reduce the cost of care to the employer/employee while helping physicians stabilize eroding reimbursements.

The direct contracting thesis is solid, and the economics work in theory and in practice. Employers get more predictable, controllable costs. Patients get access to high-quality physicians and centers. Physicians get paid fairly without the payer tax and administrative headaches. Everyone wins except the usual intermediaries who add expense without commensurate value.

Subscribe now

Why hasn’t the sleeping giant woken up?

What follows isn’t an argument that direct contracting doesn’t work. I think it’s one of the more plausible paths out of the current mess, and probably a better one for physicians than anything CMS is going to hand us or commercial payers and health systems are going to hand employers. Rather, this week’s article is more an exploration of the elements holding back direct contracting and keeping the giant asleep.

Barrier #1: Finding Physicians

In the ideal state, a self-insured CEO (or, more likely, a benefits manager) sits down with local independent physicians to work out a contract. Unfortunately, it’s getting harder to find local independent physicians. The majority of American physicians — 82% by one account — are now employed by health systems, corporate entities, or PE-backed platforms.

So, when an employer enters a direct contract, they’re usually negotiating with a health system whose market leverage and financial imperatives don’t necessarily align with the employer’s. Early in the direct contracting story, hospital brand and reputation were used as surrogates for quality and outcomes. Employers are wising up.

Even in direct contracting arrangements, large, well known health systems may leverage name recognition to seek higher payments. The argument is that their outcomes justify the expense. In reality, brand names don’t always translate to the best outcomes. In order to make the value equation work, higher costs have to be associated with demonstrably better outcomes. That often isn’t the case.

Meanwhile, many PE-backed platforms have failed to fully embrace the potential of direct contracting — at least in MSK. The unit economics of risk-sharing agreements don’t fit neatly into a pro forma spreadsheet. Many platforms lack the sophisticated practice integration and data infrastructure necessary to ensure success in direct contracting arrangements. The ROI on direct contracting doesn’t mesh with PE’s roll up strategy and exit timelines.

A final problem, true of all employment scenarios, is how to incentivize physicians who may feel more like subcontractors than direct participants. These clinicians may not feel or function like giants.

There’s no quick, easy solution to the issue of physician consolidation. Of course, direct contracting with hospitals and health systems can and has been done successfully. In the interim, independent physician groups and employers will have to find each other. Some entrepreneurial, forward-thinking groups have already done so. DTE may be the next frontier for both employers looking to control healthcare costs and physician practices trying to remain independent in a difficult economic environment. My sense is that not enough stakeholders on either side have quite figured that out yet.

Barrier #2: Infrastructure & Intermediaries

While direct contracts can be a boon for physicians, establishing and maintaining them can be challenging. Most employers want some level of cost and quality data before they’ll engage. The truth is that not all physician groups are able to easily produce this information. They may also lack the experience or sophistication to negotiate directly. Direct contracting is a different animal than typical payer negotiations.

A direct contract isn’t a typical fee-for-service arrangement. It’s custom claims adjudication, eligibility verification, bespoke billing configuration, credentialing workflows, and risk stratification. Asking a small to medium sized physician group to administer specific benefit designs across multiple employers with multiple ASOs is a bridge too far for many. Most are struggling just to keep their heads above water with their usual book of business. The same is true on the employer side. The CEO and CFO may be oblivious beyond the large “health insurance” line item on their P&L statement, and the HR benefit manager probably feels like this isn’t in their job description. Who has the time, energy, and expertise to vet multiple provider groups with varying levels of sophistication?

There are, of course, a number of intermediaries capable of bridging the employer-provider gap in this regard. Several conveners, navigators, and enablement companies have stepped in to lower the barrier to entry and facilitate contracting. Others help with data collection and aggregation. The paradox is that many of them start to resemble the middle layers they’re supposed to replace.

It’s fair to ask whose side they’re on. To demonstrate ROI to the employer, the intermediary has to deliver meaningful provider discounts. At the same time, it may add platform or PEPM fees on top. Physicians end up dealing with another third party playing them off one another in yet another race to the bottom. Paying lip service to quality then forcing providers to compete solely on cost is disingenuous, short-sighted, bad for employees, and a sure-fire way to ensure DTE never takes off.

Barrier #3: Fragmentation on both sides

I’d wager fragmentation is the single biggest thing holding back DTE, on both the demand side and the supply side. Ultimately, it comes down to a volume game: employers have to have enough employees taking advantage of these arrangements, and physician groups have to see enough volume to make the upkeep worthwhile. In my experience, neither of those conditions has been sustainably met.

Employers have dangled carrots to entice employees into taking advantage of DTE services. Carrots include partnering with “name” institutions, covering the cost of care (including travel and lodging for domestic medical tourism), and offering stepped-up care coordination services. Even so, engagement rates aren’t what many employers would like, in part because general awareness is lacking. Who is supposed to surface and promote the benefit: the employer, the convener, the physician group, or some other third party?

When carrots fail, employers may resort to using sticks — namely, refusing to cover the cost of care if the employee goes outside of the Center of Excellence network. In the past, DTE services and COE network access were optional. That may be changing with employers mandating second opinions prior to any high-cost medical treatment and pushing employees to use established, vetted provider networks.

That’s not inherently a bad thing provided that the employee gets timely access to high-quality care in an engaging and coordinated fashion. In my experience, that doesn’t always happen. Often, by the time the benefit is triggered, the patient already has an established relationship with a local clinician and/or facility they like. This is especially problematic if claims data is used because of the processing lag. Changing treating providers midstream is awkward and less than ideal. It inconveniences the employee and puts the new treating physician in an uncomfortable spot.

Care coordination is more complicated when treatment has already taken place. Someone has to own the process of gathering records, scheduling appointments, preventing redundant or wasted treatment and testing, etc. Although interoperability rules have helped, obtaining and sharing medical records remains a bit of a mess. Doing care coordination well can be capital and labor intensive — and eat into the bottom line of whoever’s doing it.

Even if you can get a reasonable number of employees to engage in the DTE process, many employers (big and small) are simply too distributed. The US is a big place, and a 5,000-employee company often has people spread across dozens of MSAs with others working remotely. There’s no local network that covers the whole population, so you’re building a patchwork solution and administering all of it.

That’s the demand side, but the supply side is fragmented too. Different geographies have different densities of physician groups, hospitals, and health systems. Some will be willing to participate, others won’t. Fragmentation can occur within a practice or health system, too. Contracting at the group level treats all physicians within that group as equal, including their costs and quality when, in truth, wide variation can exist. Getting granular to the individual NPI level requires more data and more work and can create an awkward, confusing situation.

Neither side sees enough volume to make an impact. A regional employer with 150 covered lives in a mid-sized city cannot extract meaningful concessions from a dominant hospital system. On the flipside, a practice can’t justify deep discounts and infrastructure investment for an employer who can only guarantee a dozen episodes a year.

All these challenges narrow the top and bottom of the funnel. Instead of a firehose emptying into an Olympic-sized pool, you get a trickle dripping into a tin cup. The giant reveals itself to be nothing more than a loose collection of smaller units.

One solution to the problem is to move further upstream in the care journey. Rather than focusing primarily on high-cost episodes, DTE programs would ideally start as soon as the diagnosis is made. If you’re only engaging because a back pain patient who’s already been through weeks, months, or even years of low value treatment is being offered spine fusion, it’s probably too late. Trying to steer a patient already scheduled for chemotherapy or XRT to a low-cost center is uncomfortable at best.

CMS is nibbling around this with CARA, the specialist risk provision inside the new LEAD Model, where the ACO negotiates episode prices with specialists much like an employer would. CMS itself serves as a third party of sorts, assisting with episode construction and reconciliation. Commercial payers have been slower on the uptake but are coming around to the idea of risk-sharing, bundled payments, and condition-specific episodes.

The MSK Example

Early attempts at MSK tried to solve the problem by either building around the specialist (rather than with them) or establishing “name” COE networks. In the former scenario, members got a home exercise app and care navigator while employers were sold on surgical avoidance. In the latter, brand reputation was used as a surrogate for quality, even if procedures cost more and outcomes weren’t necessarily better.

Helping patients avoid unnecessary surgery and find high-value surgeons are noble pursuits. But circumventing local specialists and using brand name as a surrogate for value has its limitations. When it comes to MSK, I think employers are catching on and want full-service solutions that incorporate high-value surgeons and facilities with regional rather than national footprints. Achieving regional density is possible but requires work on both the demand side (better employee throughput) and supply side (aggregated provider groups). Meanwhile, a digital health app only gets you so far. US News & World Report rankings, advertising spend, and pro sports sponsorships are a reflection of coffer depth not surgeon ability.

DTE is simply another flavor of value-based care, and MSK is no stranger to VBC mechanics. Orthopedic Surgeons are facing massive reimbursement headwinds, especially when it comes to the high-cost procedures employers care most about: joint replacements and spine fusions. Frankly, DTE MSK care is a natural fit for both employers and Orthopedists trying to survive in an economically challenging environment — a giant among giants.

Waking the Giant

Direct-to-employer makes a lot of sense. It gives employers more control over their healthcare spend, circumvents parts of the machinery that drive up costs while offering little in return, and has potential as an economic path forward for providers.

The giant is stirring but hasn’t fully woken up. First, we have to address the things keeping it asleep: lack of physician partners, rudimentary infrastructure, the mixed incentives of intermediaries, and demand/supply side fragmentation. These are operational and executional problems, not conceptual ones. Operational problems get solved by people who understand the operations and have the knowledge and will to fix the failure points.

Until then, DTE risks becoming VBC — a healthcare future that’s always on the way but never gets here.

BS

The problem is that the ROI on “cool tools” gets harder to defend every year. Reimbursements continue their downward march, and inflation means the cost of keeping the lights on continues to go up. Innovation in the operating room often sits as an expense item that simply flows (negatively) to the bottom line. In the current climate, practices have to take a hard look at their tech investments and the tools they adopt.

AI for the boring stuff — practice management — provides ROI you can actually measure.

Subscribe now

State-of-Play

Ambient scribes are commoditized. Practices that haven’t adopted one probably will, and it won’t matter who built it. Epic and athenahealth are shipping native versions, and startups are realizing that competing on price for a commodity is a race to the bottom. That’s a consequence of how good the foundational models have gotten and how relatively easy it is to build the scribes.

The companies creating and deploying AI scribes know this and are rapidly moving to expand their capabilities. The problem is that those capabilities are likely just as easy to commoditize. Coding, prior auth, call center automation, data tracking, and back-office workflow are useful, sensible add-ins, but they’re increasingly undifferentiated, too.

Implementations have to be not only specialty-specific but also practice environment-specific. Independent practices are facing downward pressure on revenues (declining reimbursement) and rising expenses (inflation). Profit = Revenue - Expenses is Basic Business 101. That reality is more pressing than adopting the newest OR gadget. Any successful tool must address both inputs simultaneously and do so in a way that delivers demonstrable ROI from Day One.

Whoever captures the Ortho market is going to have to do a few things: build good products, deploy quickly but carefully, and (most importantly) understand the customer.

The Actual Opportunities

Most orthopedic practices struggle with various aspects of practice management: coding levels, payer-specific prior authorization rules, denial patterns, claims submission, call center volume, referral routing — the list goes on. There are an infinite number of RCM companies, but somehow collections remain a pain point for many practices.

(Note: Please don’t DM me with your RCM solution. Most of you have already).

For a lot of surgeons, the day-to-day minutiae of practice management isn’t that interesting. The assumption is that, as long as you’re doing the work and your paycheck gets deposited, everything’s fine. It often isn’t, and continuing to downplay process inefficiencies and RCM failures is a mistake, especially given today’s practice management challenges.

With the CY2027 PFS proposing another 8–9% cut, it’s critical to stop the money you’re already entitled to from leaking out through denials, code mismatches, and accounts receivables that stretch into “bad debt” territory. At the same time, existing staff is already stretched thin. Asking them to do more or disrupting their established workflows is a recipe for attrition (which has its own costs).

This is where AI is most immediately useful, both to protect revenue (capture more of the value you’re creating) and reduce expenses (make better use of staff, consolidate and automate functions). The tools to do so already exist. Specialty-native CPT logic, payer-specific claims formatting, denial-pattern recognition, agentic prior auth, front-desk and scheduling automation are all valuable…and mostly undifferentiated.

The key is understanding the highest yield Ortho-specific use cases: PT/imaging/surgery booking, perioperative orchestration, office visit scheduling at the subspecialist level, multi-facility care coordination, ASC integration, billing/diagnosis code complexity, etc. Many busy Orthopedic practices have back offices that function on processes that evolved over decades…and people who are very attached to those processes. To outsiders, they may seem illogical. To staff, they’re all but immutable.

Things get more complicated at the MSO or supergroup level. Take the complexity of rolling out an AI tool at one practice and multiply it by the number of practices on the platform. Selling into these platforms is instant market share and an immediate leg up. But it’s not easy.

AI + VBC

Hospital-focused MSK VBC episodes already exist under TEAM with CJR-X proposed to extend that nationally. ASM is in the pipeline, and ASCs are certainly on CMS/CMMI’s radar. Love it or hate it, it’s a mistake for independent practices to ignore VBC.

Orthopedic surgeons have an uncomfortable relationship with VBC based on experiences in BPCI. But risk-bearing contracts could be a way forward in a challenging reimbursement environment. Properly structured upside can make downside worth considering. The key to favorable contracting is proving that you deliver high-quality care. Saying so isn’t enough.

None of it works unless you can capture patient-reported data at scale, cheaply, and reliably. Historically that has been expensive and resource-intensive with low response rates. What data you do collect can be messy and difficult to parse without complicated dashboards. AI can change all that.

Agents designed for autonomous, asynchronous, multichannel capture cost a fraction of the paper-and-phone-call approach with targeted, personalized outreach. The same platform that coordinates office visits, surgeries, and patient engagement can aggregate the relevant structured and unstructured data it already accesses. The right tool makes it straightforward to develop real time insights at the individual, practice, and platform levels. You’re now better equipped to enter favorable risk-bearing arrangements and to be successful once you do.

At the Ortho practice level, these are the opportunities. Ambient scribes don’t move the needle anymore, everyone knows this. The answer is scale and market capture, not creating more easily commoditized offerings. Understanding pain points is necessary but not sufficient. You must also understand how your solutions solve them in a bespoke-for-MSK way.

Caveats

AI could make care more expensive, not less.

On the revenue side, the same tools that stop legitimate leakage can be used to upcode or game prior authorization. Payers will build the same tools to fight back, leading to “the battle of the bots.” This is already happening, and the AI arms race consumes real money while patients get caught in the middle.

On the expense side, the ROI case for most of these tools is still unproven, and they aren’t free. Foundational model token costs are a fascinating topic of discussion beyond the scope of this piece (and my expertise). An AI tool that costs more than it captures just erodes margins further. The difference is that you can audit this. Denial rate, days in A/R, prior auth turnaround, staff hours per encounter, etc. can all be measured by a practice before and after AI implementation. It’s a lot harder to figure out how a robot improves your bottom line.

Regulations haven’t caught up to the pace of AI innovation, a friction point that’s playing out through AI-powered prior auth, not to mention issues around liability. I think these are near-term problems rather than long-term barriers, but that doesn’t make them any less important. There’s a significant mismatch between the pace of technological innovation and healthcare adoption that conflicts with the speed at which entrants have to capture market share to remain competitive. It’s like running a speedboat race in pitch black, uncharted waters.

You better have an experienced first mate who understands these issues and how to navigate them, someone who can credibly connect the dots between policy, economics and practice management.

Selling to Ortho

Device manufacturers and implant vendors figured out long ago how to sell surgeons a new total hip stem or rotator cuff suture anchor. Maintain a stable of “experts” who champion your product through speakers’ bureaus, instructional courses, and published research. Getting Ortho Surgeons to adopt an AI practice management tool ain’t that.

Selling a tool that changes how an entire practice operates — documentation, coding, outcomes capture, staff workflow, patient communication — is a different animal. It requires understanding what’s genuinely unique about orthopedic workflow, how to engage the staff who’ll actually use it daily and not just the surgeon who signs, and deploying fast enough to matter without breaking the practice in the process.

That’s a harder sell than a better implant. Having a bench of surgeons doing cold outreach and fluffy “thought leadership” is less important than having someone who understands the challenges practices are facing and can support sales and implementation teams with earned expertise.

Companies should be going narrow and deep with clinical advisors, not shallow and wide. Ortho Surgeons don’t want some random surgeon they don’t know telling them how great your undifferentiated AI platform is, they want someone who understands what they need and can translate it to product teams quickly and accurately.

Here’s the thing: most Orthopedic Surgeons aren’t aware that HLTH, HIMSS, or ViVE exists. They don’t listen to Raising Health, and they don’t go to LinkedIn looking for promoted posts on AI practice management tools. They want something that addresses the issues they’re facing, someone at the company they can talk to who understands those issues, and a clinical face they can trust.

They don’t want an Orthopedic Surgeon selling them, they want an Orthopedic Surgeon who knows how to support them once they’ve been sold.

BS

Disclaimer: I hold advisory/equity relationships with a few companies building AI practice management tools including Tennr, Mira Health, Healent, and Ease. Thoughts are my own.

I have too many email addresses and virtual calendars. It all started around 25 years ago when I signed up for my first personal MSN account, an email I still use (sparingly) to this day. Later, I had the brilliant idea to separate my personal and professional emails, so I signed up for an Outlook account. Once the internet decided that every single interaction required entering an email address, I created a “dummy” Gmail account to prevent my other inboxes from getting flooded with junk emails and solicitations — with modest success. (Looking at you, Constant Contact).

Then came work email addresses: one for my practice and one for each hospital or health system where I maintained privileges. Later, I added another Outlook account to separate my startup advising, consulting, and investing work from the others. A couple companies gave me an enterprise email address. Needless to say, it was a lot — especially considering that each account has its own calendar.

Welcome to 21st century entropy.

As a surgeon, you’re trained to be hyper focused on the task in front of you — to reach a flow state where muscle memory and instantaneous decision making are well balanced. In the startup/consulting/advisory world, multiple e-mail accounts, consulting engagements, side gigs, committee responsibilities, etc. can quickly spiral into chaos. Flexibility and an ability to multi-task are the most effective methods of (somewhat) managing the chaos. If only there were tools to make it easier.

Actually, it turns out there are quite a few. In fact, because AI has greatly lowered the virtual assistant barrier to entry, there are almost too many options. It was hard to distinguish between many of them, and it wasn’t clear how well any of them fit what I was looking for: a straightforward, automated way to track to-dos, calendars, and email threads with morning briefs and evening debriefs to keep me organized.

Being reasonably tech literate, I decided to build the thing myself. Chatting with Claude Pro, I created the framework of my own virtual assistant by describing what I wanted. The end result was a prompt I could feed into Cowork, using Notion as my front end. With constant tinkering and prompting, it slowly got smarter and moved closer and closer to my vision.

Subscribe now

This post details my process. It’s aimed at people who suspect these tools are useful but may be intimidated. I won’t claim what I built is super sophisticated or worthy of much more than a pat on the head. But, it works for me, and I hope it inspires others to give it a try for themselves. You don’t have to vibe code enterprise level software in an afternoon to make the effort worthwhile.

Ok, This Isn’t Really “Vibe Coding”

Technically, “vibe coding” is describing what you want to a computer in plain English and having it generate the software. That’s most definitely not what I did, but that’s kind of the point. The lesson here is that you can build something cool and useful without having to do any sort of real coding.

I didn’t build software from scratch. I took tools I was already using (Notion, Gmail, Google Calendar, Claude Pro) and, by having a conversation with AI, figured out how to wire them all together to create what I needed. I provided the blueprint and the tools, the LLM did the building. At one point, I started to feel my oats and entertained the idea of actually coding a Virtual Assistant program from scratch. Claude quickly disabused me of that idea, in large part by outlining how complex the process would be, highlighting the limitations of my own technical expertise, and surfacing things I hadn’t really thought about. (I wonder if it has similar conversations with technologists feeling their oats about healthcare!)

So, no coding required to do what I did. Just a long iterative process where I said “no, not like that, like this” a lot.

The Virtual Desk

This section details what I built and how I built it. Each individual’s mileage will vary, but that’s the beauty of AI. Once you get started, your path will quickly diverge into something that’s unique and personalized to your needs. This is what I did, not what you have to do. The following section provides one version of a blueprint and the suggested tools. What you build is up to you…and the AI, of course!

Connectors and Schedules

Connectors, integrations within the LLM that allow it to reach into other applications, are a big part of what makes the whole thing possible. Modern AI assistants seamlessly access the apps you already use, as long as you give them permission to read your calendar, email, etc. and the ability to read/write into a connected notes app or database (like Notion). You grant these permissions once, the same way you’d let any app connect to your Google account: a consent screen, a couple of clicks, and you’re done.

Connectors are 90% of the ballgame. Once the AI can see your calendar and inbox and can write to a database, everything else is just telling it what to do in plain English. The intelligence layer is inherent to the LLM; it’s the connectors that let it act on your information autonomously without needing to spoon feed it or manually enter data.

Coworking is the other piece of the puzzle. Not in the Adam Neumann sense, but in the sense that the LLM can run tasks on a schedule, unattended. That capability is what turns the whole system from a passive chatbot into a truly autonomous assistant that can function independently.

Piece 1: The tracker (the brain)

What it is: A single database — one row per live thread in my professional life. It’s part To-Do List, part running map of every open thread. Each row carries a few fields: which bucket it belongs to (advisory relationship, committee role, personal, and so on), whether I’m waiting on someone or they’re waiting on me, what the actual next action is, any relevant notes, and a due date. Later, I added a column with links to relevant emails for easy access.

There are multiple views in the tracker: a Board View organized by bucket, a Due/Overdue view that serves as the To-Do list, an Upcoming Meetings view, a Calendar view, and separate views for each bucket. The system provides redundancy and flexibility without being obtrusive or overly complicated. Just the way I like it.

How I made it: I simply described what I wanted: “make me a database with these buckets, and each item should track status, next action, a follow-up date, etc.” Claude built it in Notion in real time, then I tweaked. When I wanted it organized differently — say meetings separated from tasks, a view per bucket, a place for finished items to go — I simply said so, and Claude executed the changes. The whole thing was a natural conversation, not a slog through confusing configuration menus. Once I had the initial prompt, building v1 probably took about 30-45 minutes, most of it spent on small iterations.

What it does: The tracker is my go-to source of truth. Everything else in the Virtual Desk reads from it or writes to it. Using filters and labels, I can hide items that don’t require immediate attention and those that have a better home somewhere else in the Virtual Desk. Having one authoritative place means nothing lives only in my head, buried in one of six inboxes, or floating in the ether in between. I can even draft and park the text of follow up emails with reminders to hit send on the predetermined date. (Side note: I’m still iffy on AI drafted emails, but Claude can at least give me a placeholder. I’m still of the mind that, if it’s worth sending, it’s worth writing yourself).

Piece 2: The briefs (the daily engine)

What they are: Two scheduled tasks that run automatically — one every morning, one every evening. The morning brief reads my calendar, my last 24 hours of email, and the tracker, then writes me a short digest: here’s your day, here’s what needs attention, here’s what came in overnight. The evening debrief closes the loop — what moved during the day, what’s still open, and what should be added to the tracker. On Sunday mornings it writes a “week ahead” instead: a preview of the coming week, laid out day by day.

How I made them: A “scheduled task” is just a saved Cowork instruction (prompt) set to run at a prescribed time. I wrote out what I wanted the morning brief to contain, section by section (“start with a one-line overview, then today’s meetings, then anything needing my attention grouped by bucket, then a scan of new email”). Then I told the assistant to run that instruction every morning at 7am. That’s it. The “prompt” is really just a description of the output I want. If the output isn’t right, I tell Claude what went wrong, and it’s fixed on the next run.

What they do: The morning brief is an easily digestible look at the day ahead while the evening debrief wraps everything up. I no longer have to assemble my day from scattered sources; it’s assembled and waiting for me first thing. And because the brief reads the tracker too, it surfaces the follow-ups I might otherwise forget: “you were supposed to respond to this email today and didn’t.”

Piece 3: The weekly sweep (the safety net)

What it is: A once-a-week scan of my inbox that suggests what’s safe to archive, flags anything I may have overlooked, and catches threads that should’ve become tracked items but slipped through.

How I made it: You guessed it, a plain description of what I wanted, set to run weekly. While I could automate email archiving, I chose to have the weekly sweep just make suggestions. I’m not quite ready to completely automate this step; instead, Claude hands me a list, and I decide what to actually archive.

What it does: The weekly sweep is the backstop for stuff that quietly goes stale or gets stuck in purgatory. Once a week it forces a reckoning with the inbox and the tracker together, so nothing rots in the corner.

Trust, but Verify

The automated briefs generate, but they only act based on what’s written into the System Doc. Every morning and evening, I open a chat and have the assistant walk me through the brief one item at a time. It shows me an item; I say “mark that done,” or “push that to Friday,” or “add a note that they’re waiting on me”; it updates the tracker live; next item.

The process serves as both a review and a bias to action. The Virtual Assistant does the reading, the sorting, the drafting, i.e., the boring part. I do the deciding, the part that actually requires orchestration. That division of labor is the entire ethos of the thing. As the tool has evolved, I’ve gotten more comfortable with automating certain aspects such as generating new tracker rows based on email threads. It’s a trial-and-error process that keeps evolving.

The System Doc

Because this system changed almost daily at first, I had the assistant keep an operating manual — one document recording every rule and design decision, updated as we went. This is the source document upon which the entire Virtual Desk is based. Three weeks and forty small tweaks in, the version in my head and the version actually running had started to drift. The System Doc is the ground truth that keeps things from breaking. If you build something that evolves, remember to build the thing that remembers how it’s supposed to work.

No changes are permanent until they’re written into the System Doc. This allows for tweaks while ensuring that things don’t become irrevocably broken. It’s stored in Notion to create a layer of separation, but Claude can write into it to make necessary alterations. Should something stop working, I simply cut and paste from the Notion System Doc into Cowork. If the whole thing explodes, I can easily start over.

A little trick: the weekly sweep now reads its own manual and checks whether it’s drifted out of sync with what’s actually running. A rule I’d added in conversation was live and working, but I’d forgotten to write it down. The converse can happen too — a rule that’s in the System Doc but isn’t activating. Having the system flag its own drift makes the manual part of the machine and creates another layer of checks and balances.

The Fun Part: Tinkering

The system wasn’t completely reliable right away. It became reliable by getting things subtly wrong, me catching it, and us patching the specific hole each time. Iterating is the fun part — catching imperfections, figuring out how to do things more efficiently, clarifying instructions. Claude will mostly do what you tell it to do, the key is translating what’s in your head into unambiguous instructions. That didn’t always happen.

For example:

1. It confidently told me it was Thursday when it was Friday. Early on during testing, it built an entire daily plan around meetings that had already happened because it had the date wrong. These tools are superb at synthesis, but they can be weirdly unreliable at things you’d assume are trivial. Again, trust but verify. The problem was solved by creating standing orders to verify dates, days of the week, etc., against sources rather than assumptions.

2. It surfaced emails I’d already dealt with as if they were brand new. The cause was a vestige of ambiguity: when it searched my email for “anything from the last 24 hours,” it was searching my entire mail history not just the inbox. So, a promotional email I’d deleted and a request I’d closed a week earlier both came back flagged as fresh and urgent. The fix was one narrow instruction (search the inbox specifically, not the archive).

3. It read the tops of my email threads and missed the bottoms. On long email chains, the tool it used to scan my inbox only showed it roughly the first few messages (i.e., the oldest ones) while hiding the most recent replies. On a couple of long-running threads, it overlooked new replies stashed at the bottom of a long chain. Now, for any thread it flags, it’s required to go pull the complete thread before concluding there’s nothing new.

4. It flagged a task as overdue that I’d already finished. This one’s subtle, the kind of bug that arises from a reasonable sounding rule. I sometimes complete work directly in another system, leaving no email trail. When I finish such a task, I mark it done in the tracker. But the morning brief, checking for overdue items, was only looking at my active items. A case I’d already completed and marked done got flagged as overdue because the check that went looking for it had been told to ignore completed things. The fix: before flagging anything as missing or overdue, search everything, including the done pile.

These were all mostly failures of ambiguity, not fundamental flaws in the tools themselves. Surfacing and fixing them is the fun part. Version 1 wasn’t perfect, but the system is a lot more trustworthy now. Each misfire required a small tweak, a more descriptive rule, or a slight change in process. The practical takeaway is to expect this will happen. Treat the failures as a necessary, if not critically important, part of the process.

Quick Start Guide

If you want to build your own, here’s the stack and a suggested approach. You don’t need, and probably don’t want, my exact setup. But this should get you started.

The Tools

• Notion — For the tracker (the database) and the operating-manual doc. Free tier is plenty. Just install the connector in Claude.

• Gmail and Google Calendar — The inbox and schedule the system reads, each requiring activation of a Claude connector. Also free.

• Claude Pro, with Cowork — The AI that connects to all of the above, writes the briefs, and runs them on a schedule, unattended. $20/month.

None of these are hardwired brand choices. Any database app, your real calendar and email, and any AI assistant that can connect to them and run scheduled tasks will do, with an important caveat — make sure the apps you choose have corresponding connectors. Otherwise, automating the system can be much more complex, if not impossible.

The Build

1. Connect the AI to your calendar, email, and notes app. One-time permissions, a few clicks each. This is the step that makes everything else possible.

2. Have it build the tracker. Describe your buckets and the fields you want; let it create the database for you. Don’t overthink the structure (it’s going to change) but do be relatively detailed.

3. Seed it. Point it at your inbox and have it draft the first rows from what’s actually live. Correct what it gets wrong. Now the map exists.

4. Write the briefs. Describe, in plain language, what you want the morning and evening digests to contain. Set them to run on a schedule. Run each once by hand first and fix the wording before you trust it.

5. Add the weekly sweep. Same move — a weekly inbox scan that suggests what to archive and flags loose ends. It suggests; you decide.

6. Set the guardrails from day one. Reads freely; drafts but never sends; flags but never acts on judgment calls. Loosen only what proves reliable, only where mistakes are cheap.

7. Keep an operating doc. One page recording how it all works, updated as you tweak. Your three-weeks-from-now self will need it.

Start with just step 1 through 3 and a single morning brief. Live with that for a few days before adding the rest. The whole thing is modular. It doesn’t have to be built all at once (and probably shouldn’t be).

Conclusion

I’m a physician, not a programmer. With a little initiative, I managed to build a genuinely useful personal assistant with a few tools and a lot of patience. The barrier to doing this stuff has lowered significantly, thanks to some pretty cool tech.

If there’s a deeper lesson, it’s that there are plenty of docs like me out there curious about AI. Their comfort level and technical expertise vary widely. If you’re building AI tools for healthcare, they’re your customers, too. The exercise of building my own Virtual Assistant was a valuable learning experience. I’m no software engineer, but I have a greater understanding of AI tools.

The inverse would be a technologist spending time in a medical setting, gaining firsthand experience with healthcare delivery — triaging patients, shadowing clinicians, observing real world healthcare workflows. Someday, I’d love to read a Substack post about that experience.

BS

Predictably, many doctors were not impressed. Some pointed out that tomographic ultrasound already exists commercially. Others (again) highlighted the limitations and downsides of whole-body imaging, early detection, and overdiagnosis. While the medical establishment views these as rational, evidence-based objections, the tech world views them as gatekeeping and wet blanket-ism. The typical back-and-forth ensued with both sides talking past each other and very few minds changing.

It’s not the first time this debate has happened, and it won’t be the last.

The two sides clearly disagree about the difference between “we can generate more data” and “that data is actually useful.” The underpinning of Midjourney’s announcement was “get as many megabytes per second per dollar” of data about your body as possible, track it over time, and better health decisions are guaranteed. More data in = better probabilities out.

It’s an appealing idea, and I actually believe we will move in this direction — just slower and less dramatically than some think. It will be less about a trillion data points and more about finding the right mix of evidence-based treatment and personalization. In other words, the art of medicine evolved.

Subscribe now

Silicon Valley has been criticized for viewing humans as “meat computers,” an issue that has been exacerbated by the rise of AI. This trope takes a slightly different connotation in healthcare where uncertainty about a person’s health is treated as fundamentally a data problem. With enough inputs and compute, the right answer spits out the other end as a statistical probability — mathematical certainty patients will readily accept.

It doesn’t work that way for reasons that have nothing to do with massive data volumes. Much of what determines an individual patient’s outcome boils down to variability that can’t be easily modeled, even with more data. More inputs don’t improve the output of an imprecise equation. The human condition is more “black box” than “predictable system.” That’s why, if meant earnestly, Marc Andreessen’s claim that AI is “already smarter than 99.99% of doctors,” misses the mark. AI may have more knowledge, but knowledge is not expertise.

To illustrate this point, let’s consider a simple question posed in Orthopedic clinics multiple times a day:

Total knee arthroplasty (TKA) is about as reproducible and reasonably predictable as it gets in healthcare. Yet answering this question with any degree of patient-level certainty is trickier than it seems. How do you define “success” and what does “for me” mean when applying population-level data to individuals? Does more data make everything better?

It simply isn’t that simple.

Defining “Success”

You’d think defining “success” for knee replacement would be easy. Pain relieved, function restored, patient satisfied = done. In reality, orthopedics has spent years building increasingly sophisticated frameworks to measure success because there is no simple definition.

There are three distinct thresholds now in common use for patient-reported outcome measures (PROMs) after TKA: the minimal clinically important difference (MCID), the substantial clinical benefit (SCB), and the patient-acceptable symptom state (PASS).

MCID and SCB measure improvement relative to where the patient started. PASS measures whether the patient’s absolute post-op state is one they’d consider acceptable, independent of how much they improved. The three measures aren’t always concordant. A patient can meet the definition of one while missing the definition of another. Someone who started in bad enough shape can hit a large, clinically meaningful improvement (MCID, or even SCB) and still land in a post-op state they don’t consider acceptable (PASS…failed). Someone who started closer to baseline can fail to show a “substantial” improvement by the numbers while still landing somewhere they’re perfectly happy with.

So which one is “success”? There’s no consensus answer.

There’s often a gap between what the numbers say and what the patient says. A patient can hit every PROM threshold and still tell you the surgery wasn’t a success. An 85 year old household ambulator has different expectations than a 55 year old weekend warrior. To the former, being able to sleep comfortably again is a massive win, even if MCID isn’t reached. For the latter, inability to return to high impact sports can lead to dissatisfaction, even with pristine PROMs scores. Which constitutes a failure? By whose definition?

Meanwhile the broader healthcare system’s definition of success may have nothing to do with the patient’s experience at all. Under bundled payment and value-based care models, a TKA is a “success” if it comes in under a cost target and clears a quality threshold. A patient can develop a complication, have it managed well, end up satisfied and functional — and the episode can still register as a system-level failure because it cost three times the benchmark. Conversely, a case can hit every cost and quality box the system cares about while the patient privately considers it a disappointment.

These three definitions of success (patient-reported, clinical/functional, and system-financial) exist in the eye of the beholder. That they may disagree with each other is one of the most perplexing aspects of assessing healthcare outcomes.

On top of the definitional mess, the actual empirical picture is fuzzier than people assume. The “20% dissatisfaction rate” cited constantly for TKA traces back largely to a single 2010 study. The number gets repeated so often, it’s treated as incontrovertible fact — one that’s often trotted out to make claims about overuse and inappropriate surgery.

More recent systematic reviews peg the dissatisfaction rate closer to 7-10%. Based on the available data, the most honest answer is that dissatisfaction rates after one of the most standardized, high-volume procedures in all of surgery falls somewhere between 5% and 20%. And that depends on the robustness of study design and methodology, how “dissatisfaction” is defined, and what population gets studied. Let’s not forget — the joint replacement literature has a fragility problem. Changing the outcome of one or two patients can completely flip a study’s conclusions.

This is not a knock on knee replacement. The procedure remains one of the highest value interventions in all of medicine. Rather, it’s a challenge to the idea that “success rate” is a fixed, knowable number. More data could help here, but that’s not a foregone conclusion. You have to know what’s important and what isn’t. Large sample sizes can make trivial, clinically insignificant differences seem relevant. Studies often contain bias or lack generalizability. Each patient truly is an n of 1.

None of this means shared decision-making is pointless, or that surgeons should shrug and say “who knows.” We can and should give patients a general sense of what to expect, grounded in their specific risk factors, expectations, and the relevant literature. But “general sense” is different from a precise number, and there are no guarantees. Experience and expertise still matter more than more data in many instances. Medicine is an art. That may be unsatisfying, but it’s the truth.

The “For Me” Problem

The “For Me” problem makes things even harder. Now we’re not just trying to figure out what counts as success, we’re also trying to predict individual outcomes from models and data sets that weren’t built to do so.

There’s an analogy here that technologists should find familiar. The AI industry has gotten fairly comfortable with black boxes. The same tech ecosystem frustrated with healthcare’s constant hedging can’t fully explain why a model reaches a particular output or reliably predict when it will hallucinate. That’s not a criticism of AI. The black box phenomenon is a widely acknowledged limitation of the current state of the technology.

Human beings are black boxes too, in ways that are just as hard to predict. There are known unknowns — for instance, how a specific patient will respond to a specific treatment or whether their expectations are reasonably achievable. Unmet expectations carry the single largest risk of dissatisfaction, more than ten times the baseline risk and bigger than any surgical or technical factor. And there are unknown unknowns: undiagnosed cardiac disease, a quiescent inflammatory condition, a psychosocial stressor even the patient doesn’t fully recognize. Outcomes after joint replacement are shaped by a real mix of biological, psychological, and social factors that interact in ways we still don’t know how to fully untangle. If you’ve replaced one knee, you’ve replaced one knee.

In the spirit of this piece, I’ll admit I’m not as facile with probability and statistics as a quant or an ML researcher. I’d be glad to hear the reverse admission from the tech side more often — that there are things about medicine, and about the individual variability of human beings, that don’t collapse neatly into probability distributions.

The Hedge

Here’s the thing: I’m actually pretty excited about the potential of technology to facilitate better decision making and more personalized care. AI can take large sets of messy data and turn them into something clinically useful. But no amount of data or technology can take the art out of medicine. Infinite data is not a substitute for experience. Infinite knowledge does not guarantee expertise. That’s what’s missing in the tech v. healthcare debate.

Both sides are comfortable with their own black boxes while simultaneously underestimating their own blind spots. When it comes to the clinical utility of more data, it simply isn’t that simple.

-BS

BS

Of course, I’m a doctor, not a historian.

If you had to identify the birthdate of modern American healthcare, you’d likely point to either 1929’s Baylor Plan that gave rise to Blue Cross or the WWII-era Stabilization Act that ushered in the era of employer-sponsored health insurance. We’re now squarely in the post-modern era — astronomical expenditure, rising consolidation, and the death of physician independence. While it may seem bleak, if you look hard enough, you begin to see the seeds of healthcare’s next phase.

Call it post-postmodernism, neoclassicism, or metamodernism — the next era of American Healthcare is coming, and with it, a Future Independence Day for doctors.

Here’s why.

Subscribe now

Site Neutrality and Site-of-Service

Last week, CMS released its 2027 OPPS rules which expanded site neutral payments to include non-contrast imaging. A similar provision was introduced last year, bringing site neutral payments to drug infusions. Medicare also eliminated the inpatient only list for many surgical procedures, paving the way for more lucrative procedures to migrate to cost-effective ASCs. Earlier in the year, CMMI introduced CJR-X, a value-based care program that goes into effect in October 2027. At that time, every Medicare joint replacement performed in an inpatient setting will fall under a risk-based payment program.

The message is clear: the economics of once lucrative inpatient care, especially for high value joint and spine surgeries, is shifting.

CMS/CMMI will almost certainly continue its march to site neutrality and cost transparency, despite hospital and health system pushback. The effect will trickle down to commercial payers, self-insured employers, and savvy patients, reshaping where care is received, how it’s paid for, and who benefits. The facility fees and massive price variations for the same procedure that drive much of healthcare’s costs will slowly become a thing of the past.

So will the economic case for owning physicians.

Reverse Consolidation

Market forces have created an unprecedented wave of physician consolidation in the post-modern healthcare era. More doctors are employed today than ever. Whether its health systems, private equity firms, or payviders, the last 10-20 years have seen a massive wave of physician group acquisitions. The thesis: control the top of the funnel, control the downstream revenue it creates.

Vertical integration is the gift that keeps on giving, especially when it comes to procedural specialities. Every “owned” total joint generates OR and imaging facility fees, professional fees, skilled nursing facility admissions, home health visits, and outpatient PT referrals. The more of them you own, the more the dollars compound.

The recent shift to office and ASC-based procedures has created a different opportunity for a different set of aggregators. Regulatory tailwinds and advances in surgical techniques and protocols created an opportunity to migrate higher acuity, more lucrative procedures to owned facilities. No longer was it necessary to own inpatient facilities (and manage their complexities) in order to benefit from the revenue these surgeries generate.

Enter private equity and payviders like Optum. The ZIRP-era of low interest rates saw a wave of specialty care acquisitions. That was then, this is now.

These platforms are struggling. Few have experienced exits — the vaunted “second bite.” Most are struggling with debt service as interest rates remain high and top-of-the-market multiples are coming back to haunt them. True integration and clinical governance never materialized, and aggregators learned a hard lesson — managing doctors is like herding cats. (At least cats can be trained to use a litter box).

I suspect we are on the verge of a very painful and very eye-opening reckoning in the PE-backed PPM market, at least in Orthopedics. Few platforms are growing. Several have gone to market but haven’t found any takers. Even continuation funds and recapitalizations have been hard to come by. And interest rates may go up before they come down again.

Walmart, Amazon, CVS, and Big Tech all realized that owning traditional healthcare delivery is a fool’s errand. Private equity firms held on a little longer but are likely now reaching the same conclusion. Even Optum, once viewed as the eventual buyer of many of these platforms, is looking to divest specialty provider groups — including its Ortho practices in New York.

Health systems may reach the same conclusion about physician ownership, too. Joint and spine cases shifting to ASCs is a major hit to revenue. Mandatory VBC brings real downside risk with it and flips vertical integration on its head. Under bundled payments, referring patients to your SNF, PT, or home health services now costs money. Without gainsharing agreements, getting physicians to change their behavior is difficult; but those further eat into your bottom line. The continued march to site neutrality will almost certainly eventually come for HOPDs, and maybe even inpatient facility fees.

That Orthopedic Surgeon that once generated $3M+ in hospital revenue doesn’t look so employable. Anecdotally, some employed surgeons are already seeing changes to their comp models with fewer dollars per RVU. Employment, once viewed as the “safe” path, won’t feel so safe in the future. Doctors will only tolerate a lower ceiling when it comes with a stable floor.

We’re also headed for a reckoning beyond the financial metrics of physician employment. Doctors are becoming less tolerant of the administrative burden, loss of autonomy, and lack of representation that comes with employment. There’s a growing trend of discontentment and move towards unionization. Hospitals and health systems may decide the aggravation isn’t worth it.

Physicians’ Future Independence Day will arise from something very boring and something very apropos for American healthcare: owning physicians simply won’t make financial sense anymore. Not for health systems, not for PE firms, not for payviders. The only physician ownership that will make sense will be physicians owning themselves.

The Future Independent Physician

While the tea leaves point to a renaissance of physician independence, there’s an important caveat: many of the headwinds facing healthcare are ownership agnostic. The conditions that led many physicians to consolidate in the first place haven’t changed. Just because things may be worsening for consolidators doesn’t mean they’re getting better for independent physicians.

Site neutrality doesn’t guarantee a continued upward trend for ASC and office-based payments. Once a new market is established, old reimbursement trends may quickly re-establish themselves, leading to compressed ASC margins. Downward pressure on professional fees is likely to continue, too. Administrative burden and complex “pay for performance” mechanisms may make the specter of independent practice unattractive. To mix metaphors, the brown grass you know is better than the green grass that may not exist.

In short, the future of independent practice will look different than the past out of necessity. That challenge presents an opportunity. Employed physicians are looking for a safe landing spot — if only someone will create it.

That someone could be a national physician-owned MSO or supergroup that provides the infrastructure, negotiating leverage, and economies of scale to support indepedence in the metamodern era. It could be tech-enabled micropractices leveraging AI to reduce costs and offer data-driven quality improvements to make cash pay/direct contracting a reality. Finally, it could be novel joint ventures combining strong physician governance, real clinical leadership, and non-extractive business expertise.

The structure isn’t as important as the end result — a return to independence.

E Pluribus Unum or Ex Uno Plures?

Those who forget history are doomed to repeat it, but we’re in uncharted territory when it comes to physician consolidation and independence. Still, I’m a big believer in cycles. We’ve hit the top of the employment cycle, and it’s only natural that the pendulum should swing back toward independence. This market is long overdue for a correction, and the headwinds and tailwinds are blowing in a different direction.

From many, one — or from one, many?

The future of physician independence will be a little of both. From many physicians, one common goal of supporting each other. From one trend of consolidation, to many paths toward autonomy. Independence in spirit; strength in numbers.

Happy Future Independence Day to my physician colleagues.

BS

Private equity is everywhere — including healthcare.

Whether or not it belongs there is…controversial, to say the least. As covered in previous articles, the ship has already sailed in PE-backed Orthopedic platforms. Based on conversations I’ve had since the original pieces were published, many are rapidly taking on water. Ironically, the current state of the Ortho MSO market is better suited for the traditional PE playbook than the ZIRP-era heyday.

Those 2019-2022 vintage platforms represent classic private equity mistakes of overexuberance, lack of industry-specific expertise, fiscal and operational mismanagement, and misvaluation driven by cheap capital. These stereotypical missteps create turnaround opportunities for the more successful firms. The turnaround playbook goes:

1. Buy a distressed asset at an attractive price

2. Restructure the debt

3. Create value by improving the business (i.e., grow EBITDA)

4. Exit through an IPO or sponsor-to-sponsor sale

With so many platforms stuck in long hold periods with no obvious resolution, is this turnaround scenario feasible for PE-backed Ortho MSOs? Is it possible to rescue the ship before it goes completely under? Is the wreckage salvageable?

Maybe I’ve fallen too far down the rabbit hole, but I believe it is.

Subscribe now

Distressed Assets

The Orthopedic practices inside these struggling platforms are distressed assets, not bad businesses per se. Before being acquired, most (if not all) were mature, profitable, and on solid (if not shifting) ground. They sold because they were facing strong headwinds and believed a capital backer with business expertise could help them stay competitive and independent. That thesis hasn’t panned out.

Despite their struggles, these platforms still have ASCs that generate facility fees, strong referral networks and community relationships, reasonable payer contracts, and busy (if increasingly disgruntled) surgeons. The stable core of these practices remains, it’s the capital structure and operational integration that failed. Debit service is eroding EBITDA while true platform integration and economies of scale never materialized. PE sponsors took structurally sound Ortho practices and navigated them straight into an iceberg.

The current opportunity hinges on the idea that these platforms are still comprised of salvageable practices in need of better stewardship. A potential buyer doesn’t have to fix the economics of Orthopedics or musculoskeletal care. Instead, they have to rescue a previously functioning clinical enterprise from a failed capital structure and operational mismanagement. The solution is to recapitalize correctly, bring in real management expertise, and create meaningful platform-wide synergy — all achievable goals.

This turnaround pattern has proven successful in other sectors — including brick-and-mortar retail book sales of all things. It is the basic mechanism of private equity distressed-for-control investing: good business, bad balance sheet, motivated seller, no obvious buyer. Acquire an asset trading below its true value, bring in competent operators, fix the numbers, and exit from a position of strength.

A Roll-up of Roll-ups

The original consolidation wave rolled up independent practices into larger platforms. The opportunity now is to roll-up the distressed platforms themselves.

A buyer acquires two to three distressed but reasonably salvageable orthopedic platforms at prices that reflect current reality (likely below the sum of the obligations they currently carry). Senior debt gets restructured or paid down at a negotiated discount with lenders who would rather take a haircut than own a struggling Ortho MSO. Preferred return obligations get resolved as part of the transaction. The core enterprise, including the ASCs, payer contracts, and surgeons, carries forward into a clean structure. Additional platforms are added as the roll-up of roll-ups finds its sea legs.

This is not the original PE playbook simply rewarmed. That approach bought good businesses with cheap capital at top-of-market valuations, then piled on leverage. This new playbook is the inverse, buying good businesses at a discount and unburdening them from that leverage.

The entry point is financially favorable because sellers need relief but the buyer pool is empty. A critical component is executing the consolidation thesis original sponsors fumbled — genuine back-office integration, payer contracting leverage, standardized RCM, real economies of scale across platforms. Without it, failure is almost certain.

The first wave got a few things wrong. Physician buy-in requires more than figurehead clinical governance boards with no real authority and limited strategic input. Managing a complex network of medical practices takes more than an understanding of financial models and spreadshet math. A buyer willing and able to correct these mistakes is much more likely to achieve success.

Strategic Anchor + Capital Partner

Ultimately, this approach is just run-of-the-mill private equity distressed asset investing. The Orthopedic MSO version combines sector-specific strategic anchors with a healthcare savvy capital partner. Novel joint ventures are widely regarded as the next phase to replace PE-backed PPM roll-ups, and there is already precedent for this flavor of collaboration.

The clearest example is the partnership between the Hospital for Special Surgery, General Atlantic, and ambulatory surgery company Legent. A clinically preeminent orthopedic brand (HSS) provides standards and clinical governance; a large growth-equity partner (GA) provides the balance sheet and deal expertise; a company with ASC experience provides operational expertise (Legent). For sure, this is a cleaner ASC growth play, not a complex aggregation of failing PPM platforms. But, under current market dynamics, Orthopedic MSOs could be viewed as ASC platforms with a physician services wrapper. Rolling distressed platforms up under a recognized clinical brand, capitalized by a partner with healthcare operations experience, is a natural extension of the same logic.

The HSS/GA/Legent joint venture is merely an example. The specific names aren’t as important as the concept of how the roll-up of roll-ups might be constructed. The key piece is the combination of clinical legitimacy and standards, smart capital, and sector-specific operational expertise. That combination addresses many of the prior vintage’s shortcomings. Importantly, a meaningful clinical brand affiliation answers the quality question that PE-in-medicine always raises. Smart capital knows its limitations and when to get out of the way. Proven operational expertise ensures that mistakes aren’t repeated.

One obvious concern for the clinical partner is brand dilution. A premier orthopedic name does not want its name and reputation staked to a patchwork of debt-strapped practices. Brand association would almost have to be conditional — a baseline of quality, transparency, and outcomes reporting that platforms have to meet to carry the banner. Failure to meet the standard leads to remediation or divestiture. Harsh as that may seem, it’s how a credible brand protects itself. Holding platforms to a bespoke standard also answers the criticism that consolidation degrades quality. A model where platform access hinges on demonstrable outcomes flips the narrative on PE-backed healthcare. Finally, it enables the transition toward value-based care, favorable contracting, direct-to-employer relationships, and the next evolution of sustainable independent practice.

Finding A Different Exit

None of this matters if the new capital partner doesn’t see a clear, achievable path to an eventual exit. Like it or not, any investor — PE, VC, family office, bank, UHNWI — eventually wants a return. None of them are handing out money from the goodness of their own heart. That may be a significant source of angst, but these are not charities. What if there was an exit path that restored physician autonomy, got investors a fair return, and left things no worse for the wear?

Every PE-backed MSO to date has had the same fundamental flaw from the physician’s point of view: the exit is a sale to someone else. Strategic acquirer, sponsor-to-sponsor, continuation vehicle — in all of them, ownership simply passes from one controlling entity to the next. Physicians and their equity go along for the ride, beholden to whatever the new capital structure and employment agreement dictate. The second bite, if it ever comes, simply trades one corporate overlord for another.

The model described here designs the exit at entry, in a way that returns ownership to physicians. The mechanism is an Employee Stock Ownership Plan, or ESOP.

ESOPs Have I Loved

An ESOP is a qualified retirement plan that holds employer stock on behalf of employees. For the purposes here, it functions like an internal leveraged buyout (LBO). The ESOP trust borrows money, buys the company’s stock from the selling shareholders at fair market value, and repays the loan over time out of the company’s operating cash flow. The selling owners get cash. The employees become beneficial owners of the trust over time.

In healthcare this has a potentially favorable wrinkle. Most states restrict medical practice ownership to physicians through Corporate Practice of Medicine (CPOM) regulations. ESOP transactions can use the same legal architecture PE already uses to get around this. House non-clinical operations in an MSO spun off from the professional corporation, then the ESOP buys the MSO equity. A physician ESOP exit passes legal muster, and the corporation ultimately ends up truly physician-owned.

Here’s how it could work:

Step 1: Sector-specific strategics plus a smart capital buyer acquire the distressed platforms, restructures the debt, and spends three to five years doing the operational integration the first wave fumbled

Step 2: EBITDA recovers and grows cleanly through RCM consolidation, payer leverage, supply chain optimization, ASC efficiency, etc.

Step 3: Instead of shopping the platform to the next sponsor, the buyer exits by selling the MSO to an ESOP trust controlled by physicians. The trust finances the purchase with a combination of seller notes and third-party debt and repays it from the platform’s cash flow. The capital backer can either fully exit over time or maintain a minority stake while the brand partner/MSO collect a reasonable management fee.

Under this scenario, the buyer gets fair market value as the ESOP pays the equivalent of what any other financial buyer would pay. The selling shareholder is not taking a discount to do something noble, they are getting a full, appraised, fair-market liquidity event. The exit is achievable and worth the effort. The capital partner gets its exit, then slowly fades into the background before fading away altogether.

On the physician/ESOP side, tax structure is powerful enough to significantly improve the math. An S-corporation owned by an ESOP pays no federal income tax on the ESOP’s ownership percentage. A fully ESOP-owned S-corp is effectively a federal-income-tax-free operating entity. That shield accelerates the paydown of the acquisition debt and frees cash for reinvestment and further acquisitions. In practice, the first year’s tax savings may even exceed the cost of setting up the structure. Long term, the resulting physician-owned platform is more durable. Physicians effectively buyback their practices on more favorable terms and with less personal financial risk.

In an ESOP, governance stays with physicians. “Employee ownership” doesn’t necessarily mean employees run the company by committee. Instead, the trust owns the stock, an independent trustee represents the beneficiaries, and a board sits above management. In a physician platform, the trustee committee and the board are comprised of doctors. Under this set up, physicians own and govern the practice before and after establishment of the ESOP. Young surgeons and new partners gain ownership through participation rather than having to buy in with after-tax dollars.

That last feature addresses a significant problem created by the first consolidation wave. Original deals stratified physicians with senior partners taking cash at close, maximizing their upside and minimizing their downside. Meanwhile, younger partners ate the salary scrape and with little or no equity/cash payout to show for it. An ESOP-owned platform closes that loophole. The upside accrues to the physicians actually doing the work going forward, allocated by compensation and tenure, with no buy-in required. This is the most equitable answer to the generational unfairness present in the original transactions.

An ESOP Solves the Buyback Problem

The prior piece treated physician buybacks as the exit most aligned with the MSO thesis, but one limited by access to capital, governance complexity, and low physician risk tolerance. The ESOP is the buyback’s more palatable cousin. A conventional buyback asks physicians to come up with capital and personally underwrite leverage after they’ve already been burned once. The ESOP finances the purchase against the platform’s own cash flow, uses the tax shield to service it, and doesn’t require individual surgeons to write large checks or personally guarantee debt. It solves the cash-requirement and risk-tolerance constraints that make a straight buyback unfeasible for most groups. The governance structure still requires careful consideration, but the financial barrier is dramatically lower.

The Trade-Offs

While this all sounds good, there are trade-offs. Making this work requires doing some of the things people don’t like when PE does them. Restructuring a distressed platform means divesting underperforming assets, likely to entities that aren’t controlled by physicians. It means surgeons whose productivity or quality doesn’t meet the standard are remediated or moved off the platform. It means cost-cutting through consolidation of redundant administrative functions and staff reductions — a nice way of saying “layoffs.” It means negotiating harder with payers and, in some cases, raising healthcare prices. None of that is warm and fuzzy, and none of it should be done at the expense of patient care or physician agency. It’s a delicate balance that requires strong clinical leadership with real authority.

Despite the negative association, these difficult, unpleasant decisions are not unique to private equity. They are what any competent operator does in response to shifting market forces. Independent practices across the country are currently making the same decisions, but with less certainty and lower likelihood of success. That’s the unfortunate current reality. Consolidation pressure in Orthopedics (and other specialties) is happening with PE or without it. Hard decisions are necessary, no matter who owns or runs the business.

A key distinction of the ESOP approach is that it offers a new way forward rather than more of the same. It’s increasingly clear that the current approach to PE-back Ortho MSOs did not work. The ESOP approach pulls previously successful practices out of the wreckage, pieces them back together, then hands the improved platform back to physicians.

Executed well, there’s a quality argument to be made here too. The typical critique is that treating medicine as a business degrades care. There is evidence to support this claim, though it’s not limited to private equity involvement. The same can be said of almost all forms of consolidation, no matter who the consolidator is. The exception is physician-owned models which tend to produce better cost and quality outcomes than the alternatives. The Goldilocks endpoint here is physician ownership of a healthy, independent, national MSO based on demonstrated quality standards. In an ideal world, most would feel comfortable with that outcome.

Necessary Conditions

To be clear, this is not a universal solution, and it may not fit some in the current cohort. For it to work, a few things have to be true.

Platforms have to have healthy revenues with EBITDA degraded by out-of-control capital requirements and debt service. There’s a difference between a troubled asset and a failed business. Distressed-for-control only works when the distress is financial. If the asset is toxic — physician attrition, irrecoverable debt burden, ruined market reputation — recapitalization won’t fix things.

Lenders have to be willing to take a discounted payoff. This is more plausible than it sounds; in fact, debt restructuring is a common mechanism in PE buyouts. Private credit lenders are currently under pressure and would rather extend and negotiate than foreclose. Interest rates may be going up, not coming down. A negotiated discount is preferable to an ownership stake in a medical practice you have no interest in managing.

Physicians have to trust an exit path that’s unorthodox and different from the second bite they were initially sold. Right now, healthcare ESOPs are uncommon, especially in Orthopedics. Most surgeons have never seen one and are wary of any non-traditional practice arrangement. Selling a skeptical, recently burned physician base on a structure they don’t recognize won’t be easy. But, that’s more of an education and trust problem than a financial problem.

Finally, the joint venture operator and its participants have to do the hard integration work. The whole premise here is make the value capture of consolidation attainable and sustainable. If a new buyer simply rolls up the roll-ups using the same operation and financial playbook, it will reproduce the same mistakes. If the ESOP never materializes, everyone ends up right back at square one, if not worse.

An Earnest Call to Action

I write these pieces from the fringes, but the response to the first few made clear how many people are still in the thick of these things. Surgeon inside these platforms already know whether the picture I’ve described matches their reality. The structure I’ve laid out here can work if physicians are willing to rally around it. Acting collectively may be the only viable route to an acceptable outcome. I suspect platform leadership and capital backers understand this reality too.

Consider this an open invitation. If you’re a physician in a PE-backed platform thinking about what comes next or looking to connect with others in a similar situation, feel free to reach out either through Substack (click the button below) or LinkedIn.

If you’re a capital partner, a strategic operator, a lender, a banker, or someone building MSO alternatives and trying to figure out where the opportunity is, I’m open to a connection, too. I don’t have all the answers, but I do have a broad network and a unique vantage point.

I’m genuinely interested in whether the people who’d have to build this think it’s buildable and in connecting the surgeons who’d have to live in it with the capital and operators who’d have to stand it up. The information asymmetry that defined the first wave cut both ways, as I’ve noted previously. Closing that gap is how things get resolved favorably. There is strength in numbers, and there are more of us thinking about this privately than publicly.

Share

BS

Here are a few thoughts/takeaways from the meeting:

1. VBC in Ortho: Loving it or hating it is a choice

VBC success is out there for the taking for independent practices...if they’re willing to do the work. The lowest barriers to entry are either going direct to employers or using 3rd party conveners/navigators/aggregators. A few caveats: know your data and be able to clearly delineate your value proposition on both quality and cost. IMO, it’s a mistake for practices to treat VBC as a failed experiment. The market is moving in this direction. Winners will embrace the movement and set themselves up for success rather than fighting it and running faster on the FFS treadmill.

2. ASC Ownership is a Must

ASC ownership has become the ancillary revenue opportunity for Ortho practices. The economics of PT, MRI, and (to a lesser extent) DME are becoming less favorable and defensible. Routine total joints are now a given, and I suspect we’ll see a push towards higher acuity joint and spine cases. The key is shifting these higher risk cases to the outpatient setting safely. That requires preop optimization work, robust patient education and support, dedicated/efficient staff, and reproducible protocols and care pathways.

3. Tech’s Opportunity to Support the ASC Shift

Great opportunity for tech to support the ASC transition in all phases: pre, intra, and postoperatively. Risk stratification tools can identify patients who is and isn’t appropriate for an ASC (and modifiable risk factors that can flip patients from red flags to green lights). Intra-operative tools like surgical navigation, robotics, AR, and live feedback will make more complex procedures safer. Postop, constant contact with patients through chatbots, remote monitoring, and support services ensure smooth recoveries.

4. CMS/CMMI is Missing a Big Opportunity

CMS/CMMI are missing a huge opportunity here. Until ASM lands, there is currently no model that directly engages independent physicians or captures the economic upside of lower cost surgery centers. ASCs are on CMMI’s radar, and there will almost certainly be more to come there soon. I wrote about my vision of a physician-group focused CMS VBC model (BPCI-X) a couple weeks ago. Despite a sour taste from BPCI-A, I still believe surgeons would embrace a CMS model for joint and spine care that addresses the problems of prior models. A well-constructed model allows surgeons to capture the value they create while saving money for CMS (i.e., taxpayers).

5. Insurance company games are just another reason independent practices must rely on strength in numbers

Probably not a surprise to most, but it’s always jarring and disappointing to hear about the contracting games some commercial payers play: low-balling, dragging out negotiations, forcing practices to negotiate against themselves, constantly changing the rules. It’s yet another artifact of our healthcare system that has made it increasingly difficult to compete without the scale, market power, experience, and expertise to survive. Just like ASCs are now a must have, it seems almost certain that some form of consolidation or affiliation will be required going forward, whether that’s a supergroup, MSO, PSA, or some other form of aggregation.

6. The narrative has shifted on capital backing in Ortho

From what I heard, the general consensus is that most of the 2021-2022 era platforms got it wrong. Now, things have shifted toward novel joint ventures with strong physician leadership and true collaboration where everyone stays in their lane. The money people provide the capital; the medical people design and deliver the care. While this seems straightforward, for whatever reason, it’s not what happened in the ZIRP-era vintage.

One speaker suggested that the current crop of platforms could be facing 10-year hold periods. Yikes. Another speaker suggested that surgeons become frustrated and lose interest about 5 years into a partnership. Those two things are obviously not compatible. But that’s the position many platforms currently find themselves in. One big unanswered question: what happens to them now? Can they turn things around? If so, how? I shared my answers to these questions a few weeks ago. IMO, there is nothing inherently wrong with MSOs or joint ventures. The problem has always been the execution.

7. Frank Thomas is a role model for us all

Got to meet and shake hands with White Sox legend Frank Thomas. The “Big Hurt” and I actually share something in common: we both grew up in Columbus, GA. We even attended the same school, though not at the same time (Thomas is several years older than me and left to attend a different high school). In an era of baseball that will forever be shrouded by the PED scandal, Thomas approached the game the right way. The Big Hurt wasn’t necessarily as flashy as his contemporaries, but his quiet, consistent excellence is underappreciated.

Enjoyed my time in Chicago both on a personal and professional level. I learned a lot and had some great conversations. It’s a great time to be at the intersection of MSK, VBC, ASCs, care navigation, technology, and thought leadership. Some exciting opportunities opening up.

More to come soon!

BS

The final frontier is building a specialty-focused model that allows direct physician participation but addresses and corrects the limitations of prior models. CMS can prove it doesn’t hate specialists by designing a model that gives them full accountability without making them victims of their own success or punishing complexity.

I call it BPCI-X.

Subscribe now

Sunset without a sunrise?

Until ASM launches, there is currently no CMS/CMMI VBC model that places direct accountability and control at the specialist physician level. BPCI-A, the only Medicare bundled payment program that allowed physician group practices to hold primary episode accountability, expired at the end of 2025. TEAM and the recently announced CJR-X are mandatory but designed for hospitals. Physicians can participate through gainsharing agreements, but those are voluntary and at the discretion of the participating facility. ACO LEAD routes specialist participation through an ACO intermediary via CARA. ACCESS routes chronic MSK care through technology platforms.

BPCI has sunset without a successor, despite the fact that physician group practices outperformed hospitals by $240 per episode on lower extremity joint replacement. Three-quarters of LEJR episodes under BPCI-A were attributed to physician groups, not hospitals. Surgeons drove savings through preoperative optimization and post-acute care management, decisions they influence and are clinically positioned to make better than anyone else. Excluding physicians from direct accountability risks losing specialist engagement entirely, gainsharing agreements or not.

One valid point Abe Sutton brought up in our recent conversation is that mandatory models are easier to implement through institutions than resource-constrained specialist practices. He also highlighted a well-known problem with voluntary models — self-selection bias. High performers opt in; low performers stay away. CMS could make models mandatory for physician group practices but, as yet, has chosen not to.

Specialists perform well when held accountable, but the sunset of BPCI eliminates that option. ASM is on the horizon but will only shine on congestive heart failure and low back pain episodes. Will dawn ever come from a true BPCI successor that specialists willingly enter?

Overcoming Surgeon Skepticism

“Once bitten, twice shy” is probably the best way to describe specialists’ current attitude toward value-based care. BPCI-A worked well for surgeons at first — reconciliation payments reflected the work that went into reducing costs and improving outcomes. Then came rebasing and the race to the bottom. As target prices drifted lower, participants became victims of their own success. Upside disappeared; attrition followed. For surgeons, VBC has become healthcare’s version of a rug pull.

Despite concerns, studies suggest cherry-picking and lemon-dropping weren’t widespread in BPCI. Still, the program had rudimentary risk adjustment methodologies that effectively penalized surgeons for taking on high-risk patients. The model did little to differentiate between a healthy 65-year-old and an 85-year-old with multiple medical conditions. Any Medicare claim within the 90-day episode period was paid for out of the bundle, even if it was unrelated to the index procedure.

While CJR-X expands to HOPDs, there is currently no value-based care strategy for the most efficient and fastest-growing site of care for joint replacements — ASCs. Due to reimbursement cuts, ASC ownership has gone from “nice to have” to “must have” for independent surgeons and physician groups. Omitting ASCs creates selection bias. Healthier ASC patients are excluded, skewing the pool of eligible patients toward complex inpatient cases.

(Of note, there is an RFI for physician-owned hospital participation in TEAM and ASCs are undoubtedly on CMMI’s radar.)

Any successor model has to address these shortcomings directly and comprehensively. Surgeons will only engage if the model eliminates ratchet mechanisms, incorporates meaningful risk stratification, and embraces ASCs as an eligible site of service. The financial mechanics have to be fair and transparent without penalties for unrelated medical events or patient complexity.

BPCI-X is an attempt to design something that satisfies these criteria.

Learning from BPCI/BPCI-A

BPCI had its strengths and weaknesses. Despite its flaws, the program reduced costs without negatively impacting access and, at worst, produced equivalent outcomes to FFS. CMS is using the lessons learned from BPCI as the basis for CARA episode construction within ACO LEAD. There’s no need to stop there. Those same lessons can be used to create a new specialist-facing version of the program. Here’s how BPCI-X fixes BPCI-A’s biggest flaws.

The Ratchet Effect

Target price rebasing was arguably the biggest driver of attrition in BPCI. BPCI-X addresses this with a five-year fixed benchmark based on a National Episode Price with regional adjustment. There should be no annual rebasing, and target prices should be locked at enrollment. At the five-year mark, adjustments are permitted, but downward movement is capped such that efficiency gains aren’t penalized.

For practices already at or below the national episode price going in, BPCI-X applies an upfront target price inflation, creating a meaningful savings pool even when the starting point is already strong. For historically high-cost practices, a discounted target applies — a higher bar that reflects where the real savings opportunity exists. These adjustments avoid the trap of punishing baseline high performers while ensuring low performers can’t arbitrage the starting point delta.

Before full program implementation, CMS can either provide shadow bundle data to practices ahead of time or make Year One a shadow year with full measurement but no financial stakes. Practices see exactly where they stand before committing including expected target price, upside and downside risk, and areas of opportunity. BPCI-X is voluntary, which raises the valid concern that predicted low performers will use the shadow information to avoid the program. More on that later.

Risk Stratification

Risk stratification was significantly underdeveloped in BPCI. Taking on a high-risk patient had no reliable protection against the cost variation that comes with complexity. While evidence does not support widespread cherry-picking or lemon-dropping, there may have been some adverse selection of patients. To ensure equity, any successful model must account for complexity.

In addition to a flat national episode price with regional cost adjustments, BPCI-X layers in comprehensive patient-level risk stratification: ASA class, BMI, comorbidity index, Area Deprivation Index, and Social Vulnerability Index. A surgeon operating on a medically complex population in a high-deprivation geography shouldn’t face a worse financial position than one who selects easy cases in a favorable market. Robust risk stratification maintains access while making the model more credible to surgeons. Precedent for more sophisticated risk stratification already exists within CJR-X. There’s no reason the same methodology can’t be adapted for BPCI-X.

Mandatory vs. Voluntary

My sense is that CMMI has struggled with the mandatory vs. voluntary participation question when it comes to physician-focused VBC models. The compromise is the mandatory but limited ASM model. Properly constructed, BPCI-X doesn’t need to be mandatory to work.

A well-designed model makes participation inherently more attractive than non-participation — pull rather than push. BPCI-X can be the VBC carrot that allows specialists to escape the declining payments of FFS. If the alternative is staying in fee-for-service with a 2.5 percent efficiency cut, no MIPS exemption, and no APM bonus, voluntary participation should be sufficient. Option A is to stay in FFS, absorb the efficiency cut, carry the MIPS reporting burden, and let someone else (CMS) capture the value you create. Option B is to join BPCI-X, accept episode accountability, automatically satisfy quality reporting requirements, and access the full economics described in this piece. Some practices will prefer the predictability of FFS and accept the lower reimbursement. Others will bet on themselves.

The model should make that bet worth taking. Again, this all hinges on avoiding the rug pull and poor risk stratification of original BPCI. A high-performing surgeon capturing a value-based 30–40 percent reimbursement bonus over FFS while reducing overall episode costs is a win-win. Watching that bonus erode every year is a recipe for guaranteed failure.

A Physician Group Model Across All Settings

BPCI-X is designed as a physician group accountability model that follows the surgeon wherever they operate. It ensures specialist participants have meaningful financial stakes in their patients’ outcomes regardless of care setting. The ASC track is BPCI-X’s most distinctive contribution because no existing model covers ASC-based TJA. But the model’s logic is based on the principle that, wherever the surgeon operates, they have accountability.

TEAM and CJR-X create mandatory hospital accountability for inpatient and HOPD TJA, but they leave two gaps. ASCs aren’t covered at all, and surgeon-level engagement within hospital episodes is voluntary and at the hospital’s discretion. BPCI-X fills both gaps simultaneously, running alongside the mandatory hospital programs rather than replacing them. The result is a unified accountability structure where every Medicare TJA episode, across all three settings, carries both institutional and physician-level accountability — exactly what CMS/CMMI wants.

BPCI-X is specifically designed for independent physician group practices. Hospital-employed surgeons are already within their employer’s TEAM or CJR-X accountability structure. Adding a separate BPCI-X layer would be redundant and create unnecessary compliance complexity. The model doesn’t apply to them, although it has some physician accountability and incentive features those models should adopt (e.g., required physician leadership and mandatory gainsharing).

For joint venture ASCs, the independent physician component of the JV holds EI status; facility economics follow the JV arrangement. Restricting BPCI-X to independent groups also creates a meaningful economic distinction between independence and employment at a moment when that distinction is under real pressure throughout the specialty.

Not every Medicare TJA patient is appropriate for outpatient surgery. Standard eligibility criteria — age under 75, BMI under 35–40, ASA class I or II or low-risk III, adequate social support — screen out a meaningful share of the Medicare population. A 2024 study found that after an ASC opened at a major joint replacement program, the remaining hospital-based population averaged 69.8 years versus 66.8 previously, with 50.6 percent ASA class 3 or higher versus 41.7 percent before. ASC patients are healthier by design, as they should be. That discrepancy creates a site-of-service selection problem that BPCI-X’s three-track structure closes. A surgeon cannot improve their model position by routing complex patients to the hospital. The accountability follows the surgeon, not the setting.

A reasonable estimate for Medicare primary TJA is 30 to 40 percent currently appropriate for ASC-based care, expandable toward 45 to 50 percent through preoperative optimization. As more patients shift to lower-acuity settings through that optimization, the model captures greater savings from those lower-cost episodes. Site migration generates additional upside for both Medicare and the physician group.

Together, these three tracks cover 100 percent of a surgeon’s Medicare TJA volume while minimizing site of service discordance and muddying of incentives.

Here’s how.

The BPCI-X Framework

Track A — ASC (primary accountability)

For the roughly 35 percent of a surgeon’s Medicare TJA volume performed in an ASC, the physician group practice functions as the Episode Initiator, the same role it held under BPCI-A. CMS sets a target price based on a national episode price for each procedure type with regional adjustment. Making the index independent of historical PGP performance lessens the ratchet effect.

The savings pool is substantial — roughly $4,800 per episode — because the ASC pathway eliminates most post-acute utilization by design. The physician group receives 50 percent of net savings as its NPRA. The surgeon receives 50 percent of the practice’s NPRA: approximately $1,200 per ASC episode, more than the entire professional fee.

Track B — HOPD outpatient (mandatory gainsharing)

For the roughly 20 percent of volume performed as same-day procedures in hospital outpatient departments, CJR-X applies. The hospital is the primary accountable entity, but for physicians participating in BPCI-X, gainsharing is mandated by CMS, not at the hospital’s discretion. A defined percentage of the hospital’s episode NPRA flows to the operating surgeon, with a floor payment of $150 per episode when the hospital’s NPRA is thin. The surgeon also shares downside risk, capped at a defined percentage of professional fees.

Track C — Inpatient (mandatory gainsharing)

For the remaining 45 percent of genuinely complex cases requiring inpatient care, the same mandatory gainsharing structure applies under TEAM and CJR-X. The surgeon participates in the hospital’s episode performance — upside and downside — with the same cap and floor mechanisms as Track B.

Addressing original BPCI’s biggest shortcomings fosters surgeon buy-in. Fixing the benchmark for five years and capping future downward adjustments eliminates the ratchet mechanism and race to the bottom. Proper risk stratification ensures access and ties penalties directly to the surgical episode. BPCI-X participation satisfies MIPS and MVP reporting requirements as an Advanced APM. Between MIPS administrative costs, penalty exposure up to 9 percent of Part B revenue, and the 5 percent APM bonus, the MIPS and APM value alone is worth $36,000 to $45,000 annually. Administrative simplicity alone is worth the price of participation.

Dipping Toes into Site Neutrality

Site neutrality is viewed by many as a mechanism to begin reining in facility fee discrepancies that significantly drive up the cost of care. CMS has already displayed a willingness to embrace the concept when it comes to drug infusions. Joint replacements are a great place to go next on the journey to some form of site neutrality, and BPCI-X can be the vehicle.

A patient undergoing same-day total knee replacement at a hospital outpatient department goes home the same afternoon as a patient who had the same procedure at an ASC. The clinical experience is essentially the same, the discharge destination is the same, and the post-acute pathway is the same. Medicare pays approximately $1,200 to $1,700 more in facility fees for the hospital version.

That difference exists because OPPS and the ASC payment systems use different methodologies that build disproportionate share adjustments into HOPD rates. For an elective, same-day, home-discharge total joint replacement, those adjustments unnecessarily create economic incentive to use higher-cost but clinically equivalent HOPDs. Several published analyses suggest ASC outcomes for elective TJA are equivalent to or better than HOPD outcomes, further strengthening the site neutrality case.

A phased transition to facility payment parity for same-day TJA would eliminate the financial incentive for hospitals to keep same-day cases in their outpatient departments rather than supporting ASC development, generate meaningful additional Medicare savings, and expand the BPCI-X reconciliation pool. The savings from rate convergence — roughly $1,200 to $1,700 per HOPD outpatient episode — is meaningful given the rising volume of same day joint replacements. Rate-setting is a design question CMS would have to resolve carefully to avoid another reimbursement rugpull.

The Actuarial Case

The model only works if the math is compelling for all involved. Below is the illustrative analysis across all three tracks, at 100 Medicare primary TKA cases per year. Again, numbers are intentionally conservative and meant to be illustrative rather than projections of a specific outcome. That said, they are grounded in enough plausibility to serve as the starting point of a more rigorous actuarial analysis.

The combined picture: at 100 Medicare TJA cases per year, a BPCI-X surgeon earns roughly $96,000 to $109,000 more than under current fee-for-service — not through higher professional fees, which continue to decline, but through episode accountability across all three settings. Medicare simultaneously benefits by approximately $174,000 in net annual savings per participating surgeon. Scaling that to 5,000 surgeons generates approximately $900 million in annual net savings. The increase in reimbursement essentially functions as the inflation adjustment surgeons have never received. The MEI problem is solved and the government saves money.

Incentive Alignment & Guardrails

In BPCI-X, physician group practices that own their ASC have compounding upside that doesn’t exist in any current CMMI program. They earn the episode NPRA as the accountable entity and receive the ASC facility fee distribution as owners. When the entity making the discharge destination decision also owns the facility and bears the episode risk, alignment is complete. The incentive to optimize patients for the ASC leverages facility economics and clinical quality improvement to the benefit of surgeons and CMS. Still, there should be guardrails.

A surgeon with an ownership stake in an ASC has obvious incentive to route borderline patients toward that setting. BPCI-X downside risk mitigates this to a degree. Mandatory gainsharing on inpatient cases through CJR-X or TEAM takes risk mitigation a step further and helps ensure patients end up in the most appropriate setting.

The final check occurs if a patient transfers from the ASC or gets admitted to the hospital within the 90-day window for a procedure-related complication. The associated cost lands in the BPCI-X episode reconciliation, not TEAM or CJR-X. Push a patient into the ASC who shouldn’t be there, and every dollar of that complication reduces the physician group’s NPRA directly. The model inherently disincentivizes overreach.

There should be guardrails to protect participants as well. High-cost outlier episodes trigger stop-loss protection above a defined threshold so one catastrophic case doesn’t eliminate years of NPRA gains. Claims within the same window unrelated to the procedure itself are excluded. These provisions ensure the model doesn’t punish surgeons for events beyond their clinical control and remains accessible to smaller practices.

Finally, attribution should follow the surgeon, not the institution. For ASC cases where the operating surgeon is a member of an independent BPCI-X participating PGP, the episode is attributed to the PGP, not the facility. For inpatient and HOPD cases, the hospital is Episode Initiator under TEAM and CJR-X, with mandatory gainsharing flowing to the BPCI-X surgeon. Hospital-employed surgeons participate through gainsharing only, consistent with the structure described above. No episode carries dual attribution.

BPCI-X Simple

BPCI-X as described doesn’t fully solve one of the biggest issues with most VBC programs — administrative complexity. To address this problem, BPCI-X should offer a simplified participation track — BPCI-X Simple — for practices that want meaningful accountability without the data infrastructure full episode tracking requires. The mechanics are a case rate with a warranty: the practice accepts a pre-agreed prospective payment for each qualifying procedure. If the patient requires readmission or additional related care within a defined window, the practice shares in the cost. Upside is lower than full BPCI-X, but administrative burden is substantially reduced. Smaller or less resourced practices can participate at a reasonable tradeoff.

BPCI-X Simple serves a second purpose as a learning vehicle for CMS. Variation in how simplified arrangements perform relative to full retrospective reconciliation generates useful evidence about which practice types and market conditions benefit most from each structure. Think of it as the specialist equivalent of CARA’s Max Flex option, a mechanism for physician groups to design within the model’s framework and demonstrate results.

BPCI-X Beyond: Optimization, Coordination, and the Inversion Thesis

There is a version of BPCI-X that eventually expands into a specialist-led version of ACO LEAD, making the patient an active participant in their own optimization at the moment of maximum motivation with concrete goals and a set deadline.

Consider a patient with bone-on-bone knee arthritis, a BMI of 43, an A1c of 8.1, and a pack-a-day smoking habit. Under current fee-for-service, a surgeon can either operate and absorb the complications or tell the patient to “get healthier” with no clear pathway to do so. Neither approach is good medicine nor does it lead to meaningful chronic disease improvement. BPCI-X can fill that gap.

The surgeon benefits if the patient optimizes their health enough for the ASC and its favorable NPRA. That means providing the resources for the patient to achieve a BMI under 40, an A1c under 7.5, smoking cessation, and BP control. The surgeon now has a financial incentive to refer the patient to their PCP with specific, documented goals and a target date. The patient is motivated to improve their health such that surgery is now both indicated and medically safer.

PCP Incentives

The missing piece is the PCP’s incentive. BPCI-X should include a co-management payment — modeled on the co-management code CMS created for ACCESS — payable to the PCP for documented preoperative optimization visits associated with a BPCI-X episode. The payment should be at least on par with a Level 4 established patient visit with documented goal progress communicated back to the referring surgeon. That level of payment is small relative to total episode cost but meaningful enough to prioritize a specific patient over general chronic disease management volume.

Going a step further, BPCI-X adds a PCP outcome bonus, paid out of the surgeon’s NPRA. If the patient clears the 90-day episode window without medically related complications, the PCP receives an additional payment set as a defined percentage of the episode NPRA with a guaranteed minimum base amount. This converts the co-management payment from an administrative courtesy into a genuine clinical partnership. The PCP now has financial accountability for whether the optimization actually worked, not just whether the visits occurred.

Clearance v. Optimization

Clearance and optimization are not the same thing. A standard preoperative clearance visit is a risk assessment that happens one to two weeks before surgery. The PCP documents the patient’s current medical status, stratifies their risk for surgery, and sends a note to the surgeon. The patient either shows up as-is, has their surgery canceled, or has a rushed attempt at bare-minimum risk mitigation.

True optimization is a structured 60 to 90-day program with multiple visits, measurable targets, active management of modifiable conditions, and documented progress communicated back to the BPCI-X surgeon. That program requires visits that wouldn’t otherwise occur. Standard preoperative clearance is one visit, not six. E&M coding is appropriate for the clearance encounter. It does not capture or incentivize the structured optimization program that converts an inpatient-bound patient into an ASC candidate.

Inverting the CMS/CMMI Approach

BPCI-X satisfies the PCP-specialist care coordination loop CMMI favors, but in a slightly different way than other programs. The surgeon refers; the PCP optimizes and earns the co-management bonus and quality credit; the patient buys in because safe surgical access depends on it. Medicare saves the difference between an ASC episode and an inpatient episode — roughly $9,000-$10,000 per case. Everyone wins, including the patient who is no longer smoking and has a much-improved A1c.

Research supports the persistence of presurgical health optimization. Patients who stop smoking before surgery are more likely to remain smoke-free afterward. Perioperative glycemic control improvements show similar durability. CMS is spending substantial resources on ACCESS and dedicated chronic disease prevention programs trying to move needles that a well-designed preoperative optimization protocol moves as a side effect of getting someone ready for an elective procedure. The administration is focused on whole-person care and chronic disease reduction. BPCI-X inverts the PCP gatekeeping model CMS prefers to achieve the same results.

The Specialty ACO

Eventually, this version of BPCI-X can be further expanded into a specialist-led ACO. Specialty physician groups have incentive to bring the optimization infrastructure in-house rather than routing every patient back to a PCP for each referral. A practice with the volume to justify hiring a clinical pharmacist, a certified diabetes educator, and a smoking cessation counselor can manage the optimization pathway directly. The optimization clinic becomes a chronic disease management asset embedded within a multi-specialty group. BPCI-X becomes BPCI ACO.

Surgeon Concessions

If CMMI is willing to design a model like BPCI-X that addresses the ratchet, improves risk stratification, and restores direct physician group accountability, the quid pro quo is meaningful PGP participation. Surgeons who want guaranteed payments without accountability need not apply.

If CMS is willing to commit to protected upside, surgeons must accept real risk and commit to ongoing care improvement. Consistent financial underperformance such as negative NPRA for three consecutive years above a defined threshold, triggers probation and eventual exclusion from BPCI-X. Clinical underperformance, including safety event patterns and readmission outliers, triggers faster and harder review. Surgeons who exit the program revert to fee-for-service with no NPRA access. The same model that protects surgeons from the ratchet and rewards complexity management has to mean something when performance is genuinely poor.

A surgeon willing to look honestly at their own outcomes data, accept that some variation reflects quality rather than case mix, and participate in a public reporting structure should excel in BPCI-X. It’s up to CMS/CMMI to ensure it remains a good one.

Fin

This three-part series started with a provocative question: Does CMS hate specialists? It argued that the answer is “no,” with the caveat that CMMI hasn’t figured out the best way to incentivize them. This final piece closes by proposing a new model, BPCI-X, that addresses that shortcoming.

CMS can’t afford to hate specialists — it needs them to be willing participants, no matter the model or accountability structure. Mutual admiration is achievable, but only if the right model is designed to the benefit of both.

Ben Schwartz, MD, MBA

Editor-in-Chief

A few weeks ago I published “Does CMS Hate Specialists? — an analysis of CMS/CMMI’s recent crop of value-based care models and their conspicuous lack of direct specialist accountability. I normally prefer clever titles to clickbait-y ones, but this one was perhaps a bit inflammatory. It was, of course, a rhetorical question — one that CMMI Director Abe Sutton reached out to directly to answer.

No, Sutton informed me, CMS does not, in fact, hate specialists…and he’d be happy to chat about it. How could I refuse?

Full credit; Abe was gracious and genuinely curious. He asked specifically what I thought about the Ambulatory Specialty Model and highlighted the potential of CARA’s Max Flex Option. He pointed out the challenges of incentivizing specialists, the limits of voluntary models, and the thought process behind CMMI’s current approach. Most importantly, he took the time to listen to what this Orthopedic Surgeon-turned-blogger had to say. That willingness alone says something.

The call confirmed that the people building these models aren’t indifferent to the specialist engagement problem. But they’re working within constraints that consistently produce middling outcomes. Sutton is acutely aware of past models’ failure, and he and his team are trying to learn from those failures. Based on our conversation, I think CMMI is open to specialist-led accountability.

Building on my previous article and my conversation with Abe, here’s what I think the current models miss and a well-designed model would look like.

Subscribe now

The Flawed Thesis

I’ve often argued that CMS’ newest VBC programs — ACCESS, TEAM, ACO LEAD, and CJR-X — all send the same message: direct specialist participation is not a priority. Instead, these models are intentionally designed to shift accountability and control away from specialists. Despite evidence to the contrary, CMMI is betting that the path to better outcomes and lower costs runs around, not through, specialty care.

On the surface, this makes sense. Strip away program complexity, and the core argument is that primary care should serve as the filter, diverting patients away from high-cost specialty care and governing who gets expensive procedures. In theory, this leads to less unnecessary utilization, lower costs, and better stewardship of finite resources. In theory.

The problem is that the evidence to support this position is weaker than most think. Gatekeeping reduces utilization, but the literature fails to prove it reduces the right kind of utilization. Meanwhile, the knowledge gap that makes PCP-led management of complex conditions suboptimal has been documented extensively. Building care models around primary care accountability of specialty-focused conditions may paradoxically lead to low-value care.

This piece challenges that premise, then proposes something better. Care models built on flawed premises are destined to fail. The goal is to design an approach that produces the outcomes CMS wants; one that acknowledges current care gaps and builds bridges over them rather than gates in front of them.

Does Gatekeeping Work?

It’s widely accepted that routing all patients through primary care first reduces costs. A vestige of the Health Maintenance Organization concept, this tenet is a key underpinning of VBC. Substitute digital platforms for primary care and you get ACCESS. While it sounds good in theory, does it hold up to scrutiny?

The evidence that gatekeeping leads to higher-value care is sparse and murky. Studies supporting the idea that gatekeeping reduces specialist utilization are of limited quality. When Harvard Vanguard eliminated its 25-year gatekeeping system, specialist utilization barely changed. Meanwhile, gatekeeping lowers patient satisfaction and increases administrative burden. One study even found an association between gatekeeping and lower cancer survival rates.

CMS’ new VBC models emphasize co-management and care coordination between specialists and PCPs to address these issues. But this approach may actually be counterproductive. Specialist-only ambulatory care for osteoarthritis and low back pain is associated with improved functional status at slightly higher costs than non-specialty care. Meanwhile, co-care models cost substantially more and don’t improve function.

We simply lack robust evidence to support the claim that PCP gatekeeping of specialty care consistently drives high-value care. We also consistently underestimate the negative downstream effects of delaying definitive care — increased complexity, reduced treatment effectiveness, and paradoxically higher costs. Knowing when to refer is as important as knowing when not to.

The Knowledge Gap

When it comes to musculoskeletal conditions, the knowledge gap is well-documented. More than half of practicing physicians fail standardized MSK competency examinations. Many primary care physicians report low confidence in performing a musculoskeletal physical exam. Most NPs receive limited MSK education during their training. This is a reflection of systemic educational shortcomings, not an indictment of PCPs. Medical schools and residency programs don’t spend enough time teaching evaluation and management of MSK conditions, and efforts to address these shortcomings have produced mixed results.

I know what you’re thinking: artificial intelligence to the rescue. What better use case for AI than to upskill primary care and close the specialty-care knowledge gap? I agree in principle, but I’m not sure we’re there yet. Using LLMs to shift responsibility from specialists to PCPs doesn’t solve the liability issue. It probably makes it thornier.

When it comes to AI, high-quality outputs depend on high-quality inputs — effective prompting requires an appropriate knowledge base. Real-world clinical scenarios differ from carefully constructed case vignettes. Spotting hallucinations and dubious recommendations is difficult when you don’t know what you don’t know. Comfort managing specialty conditions comes from direct experience and learned expertise, two things AI cannot give you. Faced with uncertainty, most physicians err on the side of caution — i.e., referral — AI or not.

MSK complaints represent the second most common reason patients visit a primary care physician. The evidence suggests the clinicians managing most of those patients lack the tools to do it well. Incentivizing them to take on more of this care in the name of cost savings is the other side of the low-value coin. If overutilization (unnecessary procedures) is heads, underutilization (delayed care) is tails.

Gatekeeping and the knowledge gap are at odds with each other and with the idea of value-based care. Treatment decisions should be driven by the people most qualified to make them.

Solutions

The “PCP-as-Gatekeeper” approach has its flaws, but there are ways to address some of the shortcomings while preserving the spirit of what CMS is trying to achieve. The solutions presented below are applicable to any value-based approach to specialty care looking to strike a balance between resource stewardship and timely, appropriate intervention.

1. Close the Knowledge Gap

Evidence suggests that PCPs lack the confidence and knowledge to optimally manage MSK conditions or make timely referral decisions. The most direct fix is accessing specialist expertise at or near the decision point — not downstream from it.

The tools to do this already exist. Several studies suggest that structured, non-face-to-face specialist input (e.g., eConsults and Expert Medical Opinions) is clinically appropriate in many common MSK scenarios. A meaningful portion of the referral pipeline can be handled with a structured specialist touchpoint without a formal visit. One study showed eConsults prevented unnecessary in-person orthopaedic consults 61 percent of the time. PCPs found the consults valuable 94 percent of the time, and 97 percent of them contained actionable advice.

Specialist input early in the care process and upstream of the referral decision works best to avoid low-value care. Improper imaging, three months of fruitless conservative treatment, and misplaced patient expectations are a setup for failure in the form of wasted time, money, and resources. Early structured specialist input reduces referrals, improves quality when referrals do happen, and achieves higher patient satisfaction than gatekeeping models that simply create bottlenecks.

This workflow already exists through various digital platforms, but suboptimal implementation has produced underutilization. Medicare’s existing interprofessional consultation codes pay between $18 and $74 for specialist time, depending on duration. A 20-minute written review of a complex MSK case pays $37. The same specialist seeing that patient in the office bills at three to four times that rate. Worse, the codes can’t be billed at all if the eConsult leads to a face-to-face visit within 14 days. A specialist who reviews a case, identifies a patient who needs to be seen, and schedules an appointment gets paid nothing for the triage work that most would consider a more efficient use of resources. In such cases, the 14-day rule acts as a penalty rather than an incentive.

If upstream specialist input is genuinely valuable, payment has to reflect that. Aligning eConsult reimbursement with the complexity of the clinical work would increase specialist participation and still result in net cost reduction. Eliminating the 14-day rule would help. Thoughtful value-based care model design could also “monetize” eConsults through shared savings in a capitated or condition-specific model — which brings us to an interesting wrinkle in ACO LEAD, the CARA Max Flex Option.

2. Use CARA’s Max Flex Option

Built on the principles of BPCI-A, CARA is the mechanism by which ACOs can contract with specialists for care episodes (with help from CMS). CARA offers two participation structures: the Default Approach and the Max Flexion Options. The former uses CMS-constructed episode parameters with limited customization; the latter allows ACOs to select an existing episode and customize it by specifying trigger codes, episode length, quality measures, and performance adjustment parameters. (Max Flex arrangements are subject to CMS review for clinical appropriateness and operational feasibility).

The Max Flex Option creates an interesting opportunity to design condition-specific MSK bundles tailored to certain patient populations (global subcapitation) with comprehensive care pathways and embedded surgical episodes. Such arrangements may be better suited to achieve the goal of true co-management between PCPs and specialists. Think Integrated Practice Units and a multi-disciplinary approach based on building two-way streets, not throwing up gates.

The Default Approach will undoubtedly be easier to establish and will be the preferred path for most ACOs. Implemented intelligently, Max Flex represents an interesting opportunity for ACOs to treat specialists as collaborators, not subordinates.

Global subcapitation and other forms of well-structured risk-sharing are not solely the purview of CMMI models. Such arrangements are possible at the commercial insurance, employer-focused, and direct care levels. The main barriers are will, patience, sophistication, and incentive alignment.

3. Embed Specialist Expertise

The third solution is simpler and more immediately actionable — put a specialist Medical Director inside the ACO’s (or other entity’s) operations.

Not as a consultant or “Preferred Provider” contractor but as an operational resource. Someone with the requisite knowledge to interpret the specialty care utilization data, help navigate complex referral decisions, identify variation in pre- and post-acute utilization, and translate claims signals into clinical action. Someone who knows if a 20 percent SNF discharge rate after joint replacement is good or bad and can tell if you have a case mix problem, a surgical quality problem, a post-acute contracting problem, or all of the above.

That person also fills another, less appreciated function that has equal importance: clinical education and embedded expertise. Payment model design doesn’t fix the MSK knowledge gap in primary care, easy access to specialist-level reasoning at the point of decision does. That’s exactly what happens when you put a credible specialist in the building, make them an integral part of the care team, and leverage them to build specialty infrastructure that makes sense.

4. Flip the Script

The most ambitious solution is one CMS and others haven’t yet entertained: stop treating specialists as downstream vendors in someone else’s accountability model. Let specialist groups organize as the primary accountable entity with full cost of care risk for a defined population’s MSK spend, the data access to manage it, and the clinical authority to act on it. Partner with primary care for longitudinal chronic disease management. Own the specialist utilization decisions that drive the majority of the cost.

Evidence from BPCI-A supports this approach at the episode level. In that model, preoperative optimization served as a form of chronic disease intervention and uncovered some important lessons about the interplay between primary and specialty care. For instance, patients who stop smoking before surgery are more likely to remain smoke-free and perioperative glycemic control drives lasting behavior change for patients with diabetes. Orthopedic groups could build the proper infrastructure, aggregating around value-based capabilities, positioning as CARA Preferred Providers, and developing the analytics to take on episode risk directly. They just need a model that makes it worthwhile.

CMS could create such a vehicle explicitly, a specialist-led population health model with the same structural features LEAD provides for primary care ACOs. The evidence base and precedent exist. As yet, the model does not. Until it does, one option is to use Max Flex, embed expertise inside the organization, and make sure specialists are deeply involved. Accountability is coming one way or another; the main question is whether specialists are co-designers or bystanders, inheriting whatever primary care and payers build without them.

A Few Thoughts on The Ambulatory Specialty Model

What about ASM? Hasn’t CMMI already created a mandatory specialist accountability model? Isn’t that exactly what I’m asking for?

It’s a fair question, and ASM deserves both credit and criticism.

ASM is absolutely one answer to the question “Does CMS hate specialists?” The model explicitly acknowledges that specialists need direct clinical and financial accountability. It’s mandatory, carries two-sided risk, requires collaborative care arrangements with PCPs, and provides episode-level performance data. Participants are exempt from MIPS reporting requirements for the duration of the model, meaningful admin burden relief that any successor model should replicate. All of these things make sense. Holding primary care accountable for downstream specialty costs and specialists accountable for upstream care decisions might not produce the intended results. ASM tests the alternative.

That said, ASM as designed falls short of what genuine specialist-led accountability requires and leaves some opportunities on the table.

Specialists could rightly view ASM as another hidden payment cut with CMS taking money off the top. Some practices may run the numbers and ultimately decide that absorbing any potential penalties is better than dealing with the hassle. Others may view ASM as another reason to consider opting out of Medicare altogether.

ASM initially covers only low back pain (LBP) and heart failure. Notably, lower extremity joint replacement isn’t in the model — a significant missed opportunity. While hip and knee arthritis care lends itself to episodic, standardized care pathways, LBP is a more complex condition to manage. ASM chose the conditions with the highest Medicare spend, not necessarily the conditions where specialist accountability is most likely to work.

ASM requires participating specialists to build formal Collaborative Care Agreements with PCPs, including bi-directional data sharing, co-management protocols, and closed-loop referral communication. However, the requirements are minimal by design, and the scoring structure rewards avoiding a penalty rather than collaborating well. Both parties will take the path of least resistance to compliance, chasing metrics without necessarily improving care.

The peer-relative benchmarking structure of ASM also creates an asymmetric information problem. CMS and ACOs can use episode performance scores to construct narrow networks, but specialists have no reciprocal visibility into who is steering patients away from them or why.

Closing Thoughts

To be fair, CMS doesn’t actually hate specialists. The people building these programs are smart, well-intentioned, and have the difficult task of balancing incentives and political realities. ASM is evidence they understand that specialist accountability needs to be part of the value-based care equation rather than an afterthought. CMS is tying program design to an agenda that emphasizes chronic disease prevention, whole-person care, and improving patient health without expensive interventions. That framework is applicable to specialist-led models too.

The PCP-as-gatekeeper narrative underpinning some of CMMI’s models is less certain than it appears. Routing care through primary care only reduces costs if it prevents inappropriate procedures, and the evidence that gatekeeping selectively filters unnecessary utilization is thin. The knowledge gap that makes PCP-led MSK management suboptimal is well documented and underappreciated.

Specialists — the people CMS (and others) keep trying to route around — may be the most underutilized lever for achieving exactly what everyone says they want: better chronic disease outcomes, lower total cost of care, and patients who enter and exit the system healthier than they were before.

ASM is one answer, but perhaps not the best one.

At the end of the call, Abe suggested I might want to join “the team.” I think he was just humoring me, but I’m always open to opportunities to work on hard problems. Being a blogger/critic is the easy part — anyone can do that. Putting thoughts into action is harder... and far more fruitful. As history has proven, that’s a difficult task in healthcare reform.

I’ve already got some ideas and a blueprint — what I call “BPCI-X.”

More to come.

Ben Schwartz, MD, MBA

Editor-in-Chief

The trope is that doctors are bad at business — an opinion that’s untrue and unhelpful. Physicians were able to sell their practices for favorable multiples precisely because those practices were successful small businesses. However, doctors faced significant information and expertise asymmetry when evaluating the deals.

Preferred shares, waterfall provisions, fund structure, debt mechanics — these are foreign concepts to docs but another day at the office for PE firms. Most practices hired consultants and lawyers to parse through contract terms and get the best deal. Few took the time to understand how private equity typically works.

As surgeons, we’re taught not to make the same mistake twice. Given the response to the original articles, I thought it might be worthwhile to provide some practical thoughts for physicians in PE-backed MSOs as they contemplate what comes next.

For Surgeons

Surgeons inside PE-backed Orthopedic MSOs are facing extended hold periods, unsustainable salary scrapes, and uncertain exits. Their decision tree is narrowly defined: stay and hope for a turnaround/exit or leave and find a more tenable situation. There’s no right answer or guaranteed path forward, but there are a few things surgeons can do to better inform themselves and understand their options.

Whether you decide to leave or stay, you should do so intelligently.

Understand Your Equity and Platform Financial Health

Find your equity documents and understand what they mean — vesting schedules, waterfall mechanics, equity classes, etc. Most surgeons stopped thinking at “here’s your rollover shares, they could be worth 2-3x more in a few years.” Great, but “could be” is not “absolutely will be.” As discussed in Part 1 of the series, equity could be worth 0x, even if a sale eventually happens. Now is the time to understand base, best-, and worst-case scenarios based on data and facts, not hopes and dreams.

If possible, run the math yourself. K-1s are more than obscure tax documents you upload into TurboTax or forward to your CPA. They contain important clues about platform finances (debt, income, profit/loss), your ownership stake, and your personal relationship to the business. Know what’s contained in each box and what it means. Find a tax expert or finance person you trust and go over it with them.

Your platform should be willing to share financial data regarding revenue, collections, expenses, and the all-important EBITDA number. It serves no one’s purpose to obfuscate these numbers. Lack of transparency is a red flag and bad omen for the platform’s future. Debt structure, fund mechanics, and sponsor internal metrics are probably less accessible. Still, knowing EBITDA and financial trends is enough to construct a reasonable picture of platform fiscal health. You can reverse engineer debt from the K-1. Pitchbook, Bloomberg, Bain Capital, and others publish regularly on the state of private equity deals including headwinds, tailwinds, and transaction trends.

(Disclaimer: I’m most definitely not a tax, finance, or legal expert. Neither are LLMs. Be mindful of confidentiality, too).

Talk to a Lawyer

Yes, lawyers are expensive. No, it’s not fair that they bill hourly while you got a 2.5% “efficiency cut” this year. Certainly, you had legal representation at the time of the first sale. But the “you” in this case was your practice, not “you” the individual surgeon. Depending on group dynamics, it’s possible (if not likely) that deal terms were favorable to some docs more than others. As much as you might like your partners, only you have your best interests in mind.

Money spent on a lawyer should be spent wisely — find someone who specializes in healthcare M&A deals and has experience reviewing physician documents in PE transactions. The goal is not to re-litigate the terms of the first sale nor find fault in how things were structured. That’s water under the bridge. Instead, you should seek to understand the terms enough to make an informed decision about what to do next. That includes equity documents and employment/shareholder agreements that contain important information regarding non-compete provisions, “Good Leaver” conditions, termination clauses, and other critical provisions that affect your options.

Know Your Numbers

This is true whether you decide to leave or stay. Most physicians focus on their annual compensation but have only cursory understanding of charges, collections, overhead, ASC contribution margin, and the accounting principles that determine that number.

Your salary is being scraped, but by how much compared to what you were making before the sale and could make if you left? The sunk cost fallacy is real for many, but the clock is ticking on that upfront cash payment, and the economics of staying become less and less favorable as the hold period stretches. Understanding your exact scrape allows you to make precise opportunity cost calculations. When multiplied by the years you’ve spent in the platform (and the years remaining until a hypothetical exit) that number can be eye-opening.

Orthopedic Surgeons generate significant ancillary revenue through imaging, DME, injections, PT, and ASC facility fees. An unfavorable distribution structure can turn these into a hidden salary scrape — creating value someone else captures. The “rising tide lifts all boats” argument only works if your contribution to platform EBITDA growth translates into a favorable exit for you. Otherwise, you’re buoying creditors, LPs, and GPs while your boat’s taking on water.

It’s also important to understand the cost of starting over, whether that’s founding a new practice, joining an existing one, or taking an employed position. It’s not as simple as opening shop next door and migrating your existing practice. There may be a ramp up period during which your income drops, and you’ll spend time rebuilding referral relationships, integrating into a new practice, and dealing with the repercussions of leaving the platform. For many mid- to late-career surgeons, there’s a real financial and professional cost to starting over.

Educate Yourself About Private Equity

It’s common, and often justified, to view private equity in healthcare negatively. In reality, PE is an investment vehicle and capital source that’s neither inherently good nor inherently bad. Frankly, PE’s approach isn’t much different from that of venture capital, health systems, or even certain physician private practices. These are businesses with fiduciary responsibilities and stakeholder expectations. That doesn’t mean they get a free pass, but most surgeons who transacted into PE platforms had only a cursory understanding of the model.

Cash at close, rollover equity, the promise of a second bite — it’s easy to get distracted by these things. Fund structure, LP dynamics, carried interest, IRR goals, how and when GP and physician equity holder interests diverge are equally important. Physicians aren’t necessarily bad at business, but private equity mechanics are well outside our area of expertise. It’s worth closing that knowledge gap.

There are a number of starting points; here are a few of my favorites:

1. Barbarians at the Gate by Bryan Burrough and John Helyar — A detailed account of the RJR Nabisco leveraged buyout (LBO) by KKR. The movie version, starring James Garner as F Ross Johnson and Jonathan Pryce as Henry Kravis, is one of my favorites. No finance background required, both book and movie provide insight into the culture, psychology, and mechanics of PE deals.

2. King of Capital by David Carey and John Morris — Covers Blackstone’s rise from founding to dominance and provides a direct understanding of how modern PE funds think about deal selection, fund economics, and LP relationships.

3. The Prof G Markets Podcast with Scott Galloway and Ed Elson — Not specific to private equity but often provides insight into how investment deals are structured and funded, how to parse valuation hype from reality, and how to read market signals.

4. Plunder: Private Equity’s Plan to Pillage America by Brendan Ballou — This one obviously presents a certain viewpoint, and it’s important to point out that Ballou spent time as a federal prosecutor and Special Counsel for Private Equity in the DOJ’s Anti-Trust Division. This is not an unbiased take. Still, the book provides useful insight into how PE deals can go sideways and why incentives aren’t always aligned between sponsors and the companies they fund.

There Are No Guarantees

Everyone’s situation is different, and there are no guarantees. The platform could sell the day after you leave, or it could deteriorate into a distressed sale. The grass is not always greener, and leaving an otherwise successful practice because a favorable financial outcome seems unlikely could be a mistake. Sadly, the conditions that led you to sell in the first place haven’t materially improved. Your best hedge is to know your numbers, do your research, and close the information asymmetry gap as much as possible.

Finally, there is strength in numbers. Based on the reaction I got from the original articles, there are many other physicians out there in similar situations. Orthopedic Surgeons are some of the most driven, entrepreneurial, and collegial doctors. We like to help our own, and now’s a perfect time to do so. I’m happy to facilitate.

Epilogue to the Epilogue

My articles on PE-backed Orthopedic MSOs didn’t just resonate with surgeons — they were forwarded to platform executives, PE sponsors, and people on the fringes of these deals.

I assume most platforms and their sponsors have come to the realization that information asymmetry works both ways. Physicians may not have fully understood PE mechanics, but PE firms didn’t really understand surgeons either. Financial modeling, volume growth projections, case migration analyses, and EBITDA targets work well on a spreadsheet but quickly fall apart when faced with the realities of physician behavior and change management. A thorough audit of pre-sale practice financials provides little insight into group dynamics, individual motivations, and personal incentives.

Fringe players — accountants, management consultants, capital partners, health systems, investment bankers, M&A attorneys, and those building MSO alternatives — are curious to see how things evolve. They’re trying to figure out what, if any, opportunity exists given the current state of the market. The lack of exit activity is a drag on their businesses too.

Sooner or later, something will have to give. It’s in the best interest of everyone involved to find a resolution they can live with, whether that’s a turnaround, a buyback, or a disappointing exit. As someone who left the thicket for the fringes, I have a unique vantage point. I see turbulence, but also opportunity.

Who seizes that opportunity best remains to be seen.

Subscribe now

Author Note

I’m an orthopedic surgeon and physician executive who writes about healthcare policy, physician practice economics, and MSK value-based care at The Surgeon’s Record.

I spent years at the intersection of orthopedic practice and healthcare policy — as a surgeon in private practice, an equity holder in a PE-backed MSO, a physician executive at a healthcare services startup, a leader in national MSK organizations, and now as an advisor and advocate.

I can be reached through The Surgeon’s Record or on LinkedIn.

Questions, thoughts, and comments encouraged.

Ben Schwartz, MD, MBA

Editor-in-Chief

Experienced surgeons practicing in PE-backed MSO platforms with prolonged hold times face an uncertain future. Platform economics have diverged from expectations, but the sunk cost fallacy is strong. Based on feedback from last week’s article, many surgeons are acutely aware of the current state of play. Many are trying to figure out what comes next.

Part 1 examined the structural mechanics of PE-backed orthopedic consolidation including the investment thesis, capital structure, exit constraints, and scenarios that determine how the current cycle resolves. Private equity’s initial interest in orthopedics made sense at the beginning of the cycle but was heavily condition dependent. The conditions have changed in ways that place physician common equity in a narrow or impaired position across much of the 2019-2022 vintage cohort.

This week’s piece is forward looking and seeks to answer the following questions: What should orthopedic consolidation have looked like, what does the evidence say actually works, and what does the platform that captures that opportunity look like going forward?

The answers were available all along for those willing to look beyond cheap capital and site-of-service arbitrage. The evidence for what creates durable value in orthopedic care has been accumulating for years. Value-based care models with physician leadership, direct contracting with self-insured employers, ASC-anchored episode accountability, and physician-owned infrastructure — all of it already exists. Fund timelines and investment theses prevented most PE-backed platforms from ever pursuing these avenues seriously.

Things are different now. The distressed PE cycle is creating conditions that favor a different model, one that adapts to shifting market conditions and regulatory tailwinds while fully embracing clinical, operational, and fiscal alignment. In fact, any hope of success may depend on pivoting away from a model that miscalculated the future to one that can build it.

Part 1: The Road Not Taken

The platforms built during the current cycle achieved scale through tuck-ins and rollups. What’s less clear is whether any of them achieved meaningful integration, economies of scale, operational excellence, or clinical distinction. Are they intricately woven tapestries or poorly stitched together patchworks?

In theory, MSOs create value through standardized data infrastructure, operational efficiency, demonstrated quality, and integrated MSK condition management. The appeal for surgeons was access to business acumen, growth capital, and economies of scale. Instead, value creation in the initial PE-backed orthopedic context was mostly acquisition-driven: the number of practices under the MSO umbrella, the number of surgeons on the platform, the number of ASCs in the network.

Platforms that achieved scale without integration couldn’t demonstrate differentiated value to any sophisticated buyer, payer, or employer. They put themselves on a narrow path to success that relied upon increased productivity from established surgeons, an ability to recruit and retain young surgeons, and low-friction operational improvements. This was the easiest, safest, quickest path. The problem is, it didn’t work; bigger and better opportunities were bypassed as a result.

By the beginning of the consolidation cycle, value-based care was already an established entity. BPCI-A was generating documented savings and favorable reconciliation payments to physician groups. Employers were actively seeking partners who could deliver surgical episode accountability and high-quality, cost-efficient care. Commercial bundles existed for those willing to do the arduous work of developing them. Most PE-backed platforms chose not to pursue these opportunities.

Value-based care requires upfront investment in infrastructure that takes 3-5 years to generate returns, and those returns can be lumpy and unpredictable. A PE fund on a 5-7 year hold cycle and dependent on predictable RCM has almost no incentive to make that investment. Fund timelines and VBC ROI mechanics are incompatible. The playbook was instead to add practices quickly, generate volume-driven EBITDA growth, and attempt an exit when growth targets were achieved. Fee-for-service, purely volume driven, is the fastest, most predictable path to get there.

To be fair, VBC wasn’t — and still isn’t — a sure thing. Commercial payers have been slow to offer value-based contracts that make leaving FFS worthwhile. Of the CMS/CMMI models, only BPCI allowed physician group participation. Ratchet mechanisms eventually rendered participation untenable for many. Upside and downside risk are unattractive when you’re servicing platform debt and trying to demonstrate predictable unit economics. Platforms that avoided VBC were making a reasonable bet that the risk-bearing market wouldn’t develop fast enough. That assumption isn’t necessarily wrong if you’re planning a quick exit.

The most frequently cited example of a platform that pursued VBC infrastructure seriously is Healthcare Outcomes Performance Company (HOPCo), a PE-backed MSK platform. HOPCo claims $187 million in estimated MSK savings through the Pioneer ACO program and reports taking full global risk on more than 100,000 Medicare Advantage members with significant year-over-year savings. These results are tied to CMS program participation, which provides a degree of external validation. (Note: these figures are self-reported.)

A platform that has built genuine VBC capability in Medicare programs has the advantage of capturing emerging revenue growth opportunities. Whether that capability translates to commercial market differentiation — the ability to contract directly with self-insured employers, offer bundles to commercial payers at scale, and construct condition-specific MSK episodes — is a mostly unanswered question. HOPCo’s experience suggests that PE-backed platforms can expand into VBC under the right conditions. Whether it represents the fully integrated, commercially validated, forward-thinking model likely to succeed in the next cycle is to be determined.

Part 2: Emerging Opportunities

When private equity began making its orthopedic bets, value-based musculoskeletal care was more promise than reality. BPCI-A results were initially encouraging, but ratchet mechanisms would eventually force many out of the model. Employer direct contracting was largely theoretical outside a handful of marquee programs. Commercial payers had not developed standardized episode definitions, and the data infrastructure to support population-level MSK risk management barely existed. In short, the VBC opportunity was simply too nascent.

The market has since matured in ways that deserve closer attention. BPCI-A results demonstrated the potential of physician group performance. Employer demand for direct contracting is gaining traction — the Purchaser Business Group on Health represents employers spending more than $350 billion annually on healthcare. Employer-focused care navigation companies are aggregating center of excellence networks for specialty care. Demand exists, but credible, independent, scaled provider counterparties are lacking.

At first glance, CMS and CMMI have turned away from physician-led models in favor of health system and ACO accountability. However, CMS recently opened policy doors through the FY 2027 IPPS RFI on physician-owned hospital participation in TEAM. It’s not much, but it’s something. Current CMMI leadership seems open to suggestions. The lineup of current programs lack direct physician-level incentives and have an ASC-sized hole that could be filled by an entrepreneurial MSO. More on that later.

Recent advances in technology make building a VBC-capable platform more straightforward and resource-friendly. Outcomes measurement, episode cost analytics, and care coordination tools are becoming more economically and operationally attractive thanks to AI. The opportunity that was theoretical at the start of this cycle is more tangible and executable now.

HOPCo, with caveats about self-reported results, points toward what this looks like when pursued seriously. A platform built around MSK clinical leadership, population health infrastructure, and genuine risk-bearing capability attracts payer partnerships, employer interest, and strategic relationships that undifferentiated affiliation-count platforms cannot. Whether HOPCo has fully realized the model is unverified. That the model attracts serious partners and generates real program participation is not.

The PE platforms that are now struggling to find exits were built for a market that has shifted underneath them. The next generation MSO model described in the sections that follow is built for the market forming now, one that revolves around growing employer demand, a policy environment turning toward VBC, and more easily scalable technology infrastructure.

The road not taken might still be under construction, but it might also be the only one that goes anywhere.

Part 3: The Right Platform

Defining Features

A properly constructed orthopedic MSO has certain defining features that distinguish it from current struggling platforms.

Feature #1: No fund clock

This is the foundational design difference. A platform without a traditional PE sponsor has no fund timeline, no LP liquidity expectations, and no carry structure above physician equity. The investments that take 3-5 years to generate returns — conservative care infrastructure, data platform, payer relationship development, care coordination capability — are now more rational to make. The platform can afford to build what actually works rather than what satisfies an exit narrative.

Feature #2: No preferred equity above physician common

When physicians own the equity directly without a PE preferred return sitting above their common shares, waterfall math essentially disappears. Value created by the platform flows equitably to the physicians who created it. There are no obligations to clear before physician equity has value. This arrangement is financially superior and improves governance dynamics. When physicians aren’t superseded by preferred equity, their incentives to invest in operational quality and long-term value creation are fundamentally aligned with the platform’s interests.

Feature #3: Backend consolidation for real economies of scale

There is legitimate value to a proper MSO structure. Shared RCM, favorable GPO relationships, centralized credentialing, collective payer contracting, shared data infrastructure, and benefits purchasing are all available without PE ownership. An MSO that builds genuine shared services delivers those economies to its physician partners rather than monetizing them for someone else. The administrative efficiency that PE platforms promised but rarely delivered is achievable — it just requires organizational will to build it.

Feature #4: Governance serves the practice

Governance structures must be designed explicitly to serve the greater platform good, not investor sentiment. That means transparency and shared authority on compensation structure, care standards, capital allocation, and strategic direction. Governance discipline requires balancing physician autonomy with broader platform goals. The difference between a durable platform and an expensive legal structure around a loose confederation of independent practitioners often comes down to exactly this.

The Bain 2026 Global Healthcare Private Equity Report documents that buyers, employers, and payers are now explicitly pricing this distinction — platforms that built operational sophistication are attracting strategic interest, those built primarily on affiliation count are not. [Bain 2026 Global Healthcare Private Equity Report] An MSO built around these principles is not competing to be a better PE rollup. It is competing in a different category, with a longer runway.

The Boring Stuff

The design principles above describe the ownership architecture. What actually generates economic value for physician equity holders is less glamorous (and consistently underdelivered by PE platforms more focused on acquisition than integration). This is where the immediate and tangible return on affiliation is most apparent.

Group purchasing. A platform of 20 or more surgeons has meaningful leverage with vendors that no individual practice can match. Specialty-focused GPOs and orthopedic-specific purchasing cooperatives typically produce 10-20% savings on medical and surgical supplies relative to individual practice purchasing. One of the biggest prizes is implant standardization or reference-based pricing. Most orthopedic practices run multiple implant systems. Standardizing to one or two primary systems and committing volume creates genuine manufacturer leverage and partnership opportunities favorable to both parties. For instance, implant costs represent 40-60% of total facility cost for a joint replacement episode. For a high-volume center, better implant pricing flows directly to the bottom line.

Payer contracting. When it comes to commercial contracting, scale is critical. Payers engage differently with a group representing a significant portion of their orthopedic network in a given market. At sufficient volume, platforms can negotiate multi-year agreements with annual rate escalators rather than accepting flat (or declining) fee schedules. The most valuable contract term is not the base rate but the compounding annual increase that health systems take for granted and independent practices almost never achieve. Direct employer contracting bypasses the payer intermediary entirely and can produce rates 15-25% above commercial. That premium flows directly to physician equity holders.

Self-insured benefits and captive malpractice. A platform of 20 or more physicians can self-insure its employee health benefits with stop-loss coverage for catastrophic claims, typically saving 10-20% annually relative to fully insured premiums. A sufficiently large orthopedic group can structure a captive malpractice arrangement that reduces premiums 20-30% while accumulating equity in the captive that accrues to physician owners over time. Neither is complicated to execute. Both require organizational discipline to implement and maintain — which is the unfortunate reason most physician groups never do it.

Revenue cycle management. The least glamorous item and the one with the most immediate economic impact. Industry benchmark is 95% or above with high performers reaching 98-99%; anything less than 90% is a red flag. Improving collections through centralized coding, dedicated denial management, AI-assisted prior authorization, and systematic follow-through on unpaid claims is not exciting work. Done well, it generates more incremental EBITDA per dollar invested than almost any clinical initiative. Done poorly, it means handing payers an interest-free loan.

Shared infrastructure. Centralized credentialing, IT services, HR, and billing platform licensing at group scale reduce per-physician administrative costs substantially relative to individual practice models. The less obvious item is management talent: a platform of sufficient scale can justify hiring a full-time CFO, COO, and medical director — professional management that no individual practice could support but that every serious platform requires. The alternative is physician time redirected from clinical work to administrative management, which is both economically inefficient and the primary driver of physician burnout in independent practice.

None of this is novel, health systems have been doing it for decades. There’s no reason MSOs can’t do the same. In a well-constructed platform, the savings from these initiatives flow back to the physicians who built them rather than to institutional overhead or preferred return. Doing the boring stuff well has tremendous compounding effects.

“Reverse Consolidation”

The distressed PE cycle is creating a specific opportunity for forward-looking physician groups. Call it “reverse consolidation” or re-affiliation under physician ownership — surgeons who moved from independent practice into PE platforms during the consolidation wave are now looking for an alternative structure.

This will not be a return to the practice they sold — that definition of independence has changed. Such practices were increasingly exposed to an unfavorable environment that included eroding margins, limited contracting leverage, expensive infrastructure, and market vulnerability. What a true MSO offers is the right balance between independence in ownership and consolidation in operations. Physician partners own the equity and control the governance. The MSO provides an integrated, efficient operational structure and backend that levels the playing field.

ASC assets represent a parallel opportunity. Distressed platforms with physician-owned or platform-controlled ASC infrastructure may offer acquisition opportunities at favorable valuations. A next-generation MSO that acquires well-located ASC infrastructure during this period captures facility fee economics more efficiently than PE-backed platforms could. A fresh start also nullifies prior ASC incentive structures that were unattractive to younger surgeons or high producers with limited ownership.

Surgeons becoming available through platform distress are entrepreneurial, efficient, high-quality operators who made those platforms worth building in the first place. They bring established referral networks, proven clinical skills, and deep medical community roots. Having built successful private practices prior to the sale, they have the necessary operational sophistication. Now, their firsthand knowledge of the strengths and weaknesses of the PE model makes them better equipped to build the alternative.

PitchBook’s Q1 2026 Healthcare Services Report categorizes MSK as a sector with “aged inventory” — at least 25 PE-backed companies with holding times of five or more years. The available talent pipeline is considerable. Build it, and they will come.

Alternative Capital

As discussed in Part 1, capital is the most significant obstacle to physician buybacks of distressed PE platforms. An orthopedic group repurchasing its practice needs money to fund the transition, re-establish back-office systems, and re-acquire their ASC. That requirement — typically several million dollars for a mid-sized group — has no obvious source in a world where the only efficient deployer of capital at that scale has been PE.

Family offices and other forms of “patient capital” are underexplored solutions that could enable re-affiliation under physician ownership.

Structuring the deal. Consider a distressed platform with significant senior debt, modest EBITDA, and physician common equity deeply underwater. A clean exit is not available at a price that satisfies all obligations. But a negotiated asset purchase in which physicians and a long-term capital partner acquire the operating assets may be achievable. In this scenario, purchase price reflects current fair market value rather than peak multiples, and the PE fund and lenders resolve what remains of the debt obligations.

The transaction works as follows: physicians in the platform contribute meaningful equity — say, $500K to $1M per surgeon for a group of 20 to 30, generating $10M to $30M in physician capital. A family office or closely aligned investor provides supplementary capital at non-extractive terms: $30M to $50M at an 8-10% preferred return with no carry, no mandatory exit timeline, and governance rights that preserve physician majority control. The combined capital purchases the clinical operations, physician employment agreements, ASC equity, and management systems out of the distressed platform at a negotiated price lenders accept as superior to forced resolution.

The result is a new entity with physician majority equity, a capital partner in it for the long haul, no PE preferred return above physician common shares, and no ticking fund clock. The MSO functions as it was always supposed to — with genuine operational integration, systemic value creation, and long-term physician economic alignment.

What’s in it for lenders. Private credit lenders are more interested in recovery than strict covenant enforcement. A negotiated asset sale to a physician-led entity with long-hold capital backing produces a more orderly recovery and avoids the operational destruction that accompanies a messier resolution. Lenders who understand that the alternative is one in which surgeons leave and the asset is worth a fraction of current value have a rational basis to negotiate a discounted payoff as part of a physician-led transition.

The PE fund’s reasoning is more complicated. They are absorbing a loss on preferred return in exchange for a negotiated resolution. While certainly disappointing to GPs and LPs, a controlled resolution that closes the book on a troubled asset at some recovery value is preferable to a prolonged workout that consumes management time and frustrates limited partners.

What qualifies as “patient capital.” The most realistic candidates are strategic investors with longer hold periods and lower return targets than traditional buyout funds. Patient capital accepts less outsized financial return in exchange for strategic positioning and a chance to fund meaningful healthcare reform.

Physician equity as the foundational capital source is the cleanest option and eliminates outside investor governance entirely. True family offices with healthcare backgrounds provide perpetual capital without fund termination pressure or LP return requirements. Implant manufacturers and medical device companies represent a third category — one with established precedent.

As documented in Part 1, pharmaceutical distributors have already executed large-scale acquisitions of specialty physician platforms in gastroenterology (Cardinal Health/GI Alliance), ophthalmology and retina (Cencora/Retina Consultants, McKesson/Prism Vision Group), driven by the strategic value of owning the physician relationship that controls device utilization. An orthopedic equivalent applies the same logic to a different specialty. The manufacturer or distributor accepts below-market financial returns because the strategic value of the relationship exceeds pure investment return. A physician-led MSO with proven outcomes data and meaningful volume is the right kind of partner. The Geisinger/Medacta outcomes-linked contract model discussed in Part 6 is an example of this alignment; a deeper version involving equity is conceptually interesting but would require careful AKS and Stark architecture.

Health systems represent a similarly nuanced option. A health system taking an equity or governance stake in an independent MSO creates conflicts that undermine the physician ownership thesis. But a long-term professional services agreement (PSA) under which the health system pays the MSO for clinical services, care coordination, or network participation provides capital-equivalent value without equity transfer. Payments under a well-structured PSA generate predictable revenue that funds practice operations without diluting physician ownership. The health system gets a referral relationship, direct contracting capability, and a VBC MSK partner while the physician MSO retains independence. The downside: an exclusive long-term PSA with a single health system risks recreating precisely what the model was designed to avoid.

PE platforms approaching resolution over the next 24-36 months need to consider all options. Physicians inside those platforms have three choices: wait for a health system acquisition that rarely delivers the economics they expected, wait for a sponsor-to-sponsor recap that the market has largely closed, or organize themselves with capital partners to execute the transition to physician ownership. The third path has been theoretical because the capital sourcing problem had no obvious solution. Family office and patient capital structures are that solution, but physicians and their advisors must actively seek them out instead of waiting for them to appear.

Part 4: Direct Contracting, Cash Pay, and Opting Out

The conditions that led to PE-backed orthopedic practice consolidation at the start of the current cycle have not improved. If anything, they have worsened with continued reimbursement cuts, rising practice costs, and growing administrative burdens. PE-backed platforms bet hard on ASC site-of-service arbitrage, attempted to juice RCM, and went on a hiring spree — yet they still came up short. New approaches are necessary if there’s any hope of independence. Direct contracting, cash pay, and traditional system opt-out are potential solutions.

Direct Contracting

The self-insured employer market represents more than 60% of commercially insured Americans. Orthopedics is consistently one of the highest single-category expenditures in employer health spend. Large employers have been trying to solve this conundrum for years with varying degrees of success.

The Walmart Centers of Excellence program documented $86 million in savings on spine surgery, though not how you might think. Ironically, surgeries themselves were more expensive at the COEs, averaging $32,177 compared to $29,770 at non-COE facilities. The primary cost saving mechanism was procedural avoidance in the form of second opinions that recommended against surgery. The implication is that orthopedic value creation must span the spectrum of care, not just the surgical episode. That thesis has continued to gain traction since the original 2019 HBR article detailing Walmart’s program.

Platforms with genuine conservative care pathways offer something fundamentally different from platforms that optimize for volume. The Purchaser Business Group on Health has documented consistent employer demand for exactly this model. Until now, demand has been uneven and fragmented, but that tide appears to be turning. More employers are offering care navigation benefits and mandating that employees use them. The supply-side of credible physician-led counterparties with outcomes data and risk-bearing capability remains available to early movers.

A next-generation MSO is better positioned to offer this than a health system whose revenue depends on surgical volume and market-driven procedure charges. A platform that offers a conservative care pathway and a cost-efficient ASC has fully aligned incentives — the right clinical answer for the employee is also the right financial answer for the employer.

For an MSO building toward direct employer contracting, one product is a fixed-price 90-day episode inclusive of conservative care evaluation, second opinion when clinically indicated, surgical facility, implant, PT, post-acute coordination, and an outcomes warranty. Employers get cost predictability and a quality guarantee. The platform captures revenue currently absorbed by the payer or large health system. The ability to present credible data is essential. Employers want to know outcomes by surgeon, episode cost by procedure, complication and readmission rates, and patient satisfaction scores. Without this information, direct contracting becomes another dead end.

Cash Pay and Transparent Prices

A more “radical” version of direct contracting is the fully transparent, cash-pay model pioneered by the Surgery Center of Oklahoma. Founded by anesthesiologists Keith Smith and Steve Lantier in 1997, SCO posts bundled all-inclusive prices for every procedure online, accepts no insurance contracts, and has built hundreds of direct contracts. Self-funded employers fly their employees to Oklahoma City for surgery at prices often 6-10 times lower than comparable hospital rates. The model is physician-owned, incentive-aligned, and operationally lean in ways that insured-model platforms cannot replicate.

The SCO model is a genuine proof of concept for transparent pricing and direct employer contracting. Cash pay does limit the patient population for many orthopedic services, expensive spine and joint procedures chief among them. But for the right practice mix and patient population, it represents the purest expression of what physician ownership and incentive alignment can produce.

The transparent, cash-pay model may have seemed niche when the current cycle began. But the movement is gaining traction. Beyond employer-driven direct contracting, many patients trapped in high-deductible health plans are finding that cash pay is cheaper than using their insurance. “Cash pay by default” is creating a marketplace for savvy healthcare consumers, and startup companies are seizing the moment, offering tools to support both the supply and demand sides of the market.

Medical tourism — domestic and international — represents a parallel opportunity. Domestic COE models already demonstrate employer willingness to pay for employee travel when quality and value justify it. Internationally, patients in Canada, the UK, and other single-payer systems face significant delays for joint and spine care. Platforms that differentiate on price, technology, expertise, and experience can attract cash-pay patients from broader geographic catchments than traditional private practices could reach.

Opting-Out

There are multiple ways for platforms and surgeons to opt-out of a system that is increasingly at odds with the concept of independence.

Go out-of-network. On the commercial side, going out-of-network frees surgeons from burdensome payer rules and one-sided contract negotiations — though the cost may be borne by the patient. The No Surprises Act and IDR process are beyond the scope of this piece, but some groups have used this framework effectively. A hybrid approach is to opt out at the surgeon level while keeping the ASC in-network. This lowers the barrier to entry for the patient, offsets declining professional fees, but still offers savings on facility costs.

Establish IPAs. A less radical option than commercial opt-out — and potentially more valuable for established groups — is converting from hospital-aligned PHO structures to Independent Physician Associations (IPAs). IPAs take the MSO concept and supercharge it by eliminating the hospital tax that exists within PHO structures and creating platform-wide negotiating leverage. Managing administrative costs is critical to IPA success, and separating from health system referral relationships carries short-term risk. There may be an initial volume dip, but a large, well-run MSO that delivers great outcomes and superior patient experience should be able to overcome these hurdles in time.

Opt out of Medicare. Medicare opt-out has long been the “break-glass” option for orthopedic surgeons frustrated with declining reimbursement and growing administrative burden. It removes the physician from CMS fee schedule constraints, enables direct contracting for professional fees, and eliminates mandatory quality program participation. These benefits are attractive to an MSO trying to simplify billing and improve unit economics. However, full opt-out is practically limiting for most joint replacement and many spine surgeons whose patient population is predominantly Medicare-age. The number of orthopedic surgeons who have actually done it remains small — less than 2%.

Rather than turning their back on Medicare entirely, orthopedic surgeons may be better served by engaging with CMS models that provide meaningful upside. The problem is that current models lack a clear path for direct physician group or ASC participation.

Therein lies an opportunity.

Part 5: Introducing BPCI-X

The peer-reviewed evidence from BPCI and BPCI-A is clear: physician group participants reduced 90-day episode spending more than hospitals, achieved greater readmission reductions, and drove the majority of joint replacement participation across the program. Despite this evidence, CMS has moved in the opposite direction.

TEAM, mandatory for certain hospitals beginning January 2026, and the proposed CJR-X, mandatory for every acute care hospital nationwide beginning October 2027, both make hospitals the sole accountable entity with no provision for physician group or ASC participation at the convener level. The FY 2027 IPPS Proposed Rule, released April 10, 2026, contains two RFIs that partially reverse this trajectory: one asking for feedback on ASC participation in TEAM as early as CY 2028, and one asking whether physician-owned hospitals in non-selected CBSAs should be allowed to voluntarily participate. CMS’s own language cites evidence that physician-owned hospitals help control costs, maintain outcomes, and prevent consolidation.

Instead of dedicating lobbying dollars and resources to challenge corporate practice of medicine laws, it might be time to focus attention on a potentially bigger opportunity. The RFIs in the 2027 Proposed Rule demonstrate that CMS is open to broader discussions about physician-owned hospitals and bringing physician groups back into the fold. The current administration wants innovators with bold ideas, not complainers with grievances to air. Sophisticated MSOs should treat this as an opening.

What the organized orthopedic community should be submitting before June 9 is a specific, evidence-based alternative framework — what I call BPCI-X.

The BPCI-X Framework

The goal of BPCI-X is to build on the success of prior physician-led models while incorporating constructive elements from TEAM, CJR-X, and ACO LEAD. CMS has a significant ASC and physician group gap in its current episode accountability portfolio that BPCI-X could fill. Under this model, orthopedic MSOs function like ACOs to deliver coordinated, high-quality, expert-led musculoskeletal care. It’s a win for patients, physicians, and taxpayers.

Provision 1: Physician group as eligible accountable entity. Original BPCI established that physician groups bear episode accountability effectively and outperform hospitals on surgical episode cost reduction. Any successor model should restore physician group eligibility that doesn’t exist in either TEAM or CJR-X.

Provision 2: ASC-eligible episode accountability. A model that applies only to hospital-based episodes is not a model for orthopedic care in 2026. ASC-based total joint replacement is now standard practice. An episode accountability model that excludes ASCs creates a perverse incentive to perform procedures in more expensive hospital settings to trigger episode payment participation. That is the opposite of CMS’s stated cost-reduction objective.

Provision 3: Condition-specific episode bundles. A well-governed orthopedic MSO participating in BPCI-X should be accountable for distinct condition-specific bundles starting with lower extremity osteoarthritis and low back pain. CMS has claims data tied to specific diagnosis codes. Use of that data shouldn’t be limited to surgical episodes. Condition-specific design with nested surgical bundles creates greater value and is more clinically and financially defensible.

Provision 4: Simplified data collection and reporting. One of the primary barriers to physician group participation in prior CMS episode models was the administrative complexity of quality reporting and data submission. BPCI-X should simplify reporting requirements and incorporate the least burdensome PROMs instruments. Claims-based quality measures should be used wherever possible, reducing the need for separate data submissions. New participants should have 12–24-month ramp periods before full downside risk exposure, similar to the TEAM Track approach.

Provision 5: Orthopedic MSO as ACO-equivalent risk-bearing entity. Orthopedic MSOs should be eligible to function as an ACO-equivalent structure for episode accountability. BPCI-X participants bear downside risk, manage the care pathway from conservative care evaluation through post-acute rehabilitation, retain shared savings generated through episode efficiency, and report on quality metrics that are clinically meaningful for the musculoskeletal patient population. Risk adjustment mechanics should mirror other established CMS programs.

Consolidation is a major cost driver in healthcare, and CMS has arguably made it worse with policies that disadvantage private practices. CMMI has an opportunity to turn things around by creating programs that promote independence. Current models don’t directly tackle this issue — BPCI-X does.

Part 6: Novel Models — The Frontier

The models described above are largely nascent or theoretical, but they’re also achievable. Given the current climate, they may be the only viable path to independence for orthopedic surgeons in the short term. Part 6 presents long horizon, less conventional opportunities whose timelines extend beyond the immediate planning horizon.

Platform-wide AI infrastructure. The most immediate value of AI in an orthopedic MSO revolves around administrative efficiency, operational simplicity, revenue cycle mastery, and cost-effective data handling. AI-assisted prior authorization reduces the denial rate and administrative burden that consumes physician time and degrades clinical culture. AI-assisted coding, documentation accuracy, and claims management captures revenue that currently leaks through denial.

At the ASC level, AI-driven supply chain analytics identifies implant cost variation by surgeon, procedure, and patient complexity in ways that support standardization without mandating it. These applications generate real ROI at platform scale and simultaneously build the outcomes data infrastructure that enables everything else described in this piece. Data is the contract negotiation. Building the AI infrastructure to capture it is a foundational investment.

AI may also be the tool that finally makes value-based care operationally achievable at scale. Whether through predictive analytics, enhanced patient engagement, systematic PROM capture, or intelligent program dashboards, AI has the potential to solve the administrative complexity problem that has made VBC participation expensive and burdensome for physician groups.

Outcomes-linked implant contracts. Geisinger Health System demonstrated what this looks like in practice through a partnership with implant vendor Medacta. Beginning with total hip replacement in 2018 and expanding through total knee, revision knee, lumbar spinal fusion, and reverse shoulder arthroplasty by 2024, Geisinger extended lifetime guarantees to select patients covering all complication-related care costs, shared between the health system and the manufacturer.

The institution and the manufacturer share outcome risk, and patients/employers get a degree of assurance. An MSO platform with comparable outcomes data and volume commitments occupies the same negotiating position. Alignment may be even stronger since physicians making implant selection decisions also bear the financial consequences of complications.

Implant company and distributor strategic alignment. Healthcare distributors have demonstrated appetite for vertical integration into specialty physician platforms. Cardinal Health acquired GI Alliance at a $3.9 billion valuation, Cencora acquired Retina Consultants of America, McKesson acquired Prism Vision Group. The strategic logic — owning the physician-driven product utilization relationship — is increasingly compelling across medical device categories as well.

Orthopedics presents complex legal and regulatory issues given AKS constraints around implant economics and physician selection decisions. But an MSO with transparent outcomes, scaled ASC volume, and direct employer contracting capability is an interesting strategic asset for entities seeking to own the MSK care continuum. Any such arrangement would require careful Stark Law and AKS architecture, likely a formal OIG advisory opinion, and risk tolerance from the device partner. The logic makes sense; execution requires legal sophistication.

The MSK insurance carveout. Behavioral health and pharmacy carveout models demonstrate that a defined, high-cost, actuarially predictable benefit can be managed more efficiently by a specialized entity than by a general health plan. Musculoskeletal spend has similar characteristics: predictable population demographics and conservative care pathways that reduce surgical rates without negatively impacting outcomes. A sophisticated MSO could serve as a capitated, outcomes-linked MSK benefit manager for a defined employer or payer population. This is a 5-10 year build, not a near-term play, but for a platform designed with physician ownership and a long time horizon, it’s a logical destination.

Wrapping Up: Something for Everyone

Diagnosing the problem, as in last week’s article, is the easy part. This article presents a way forward — one that’s not easy but certainly possible. The infrastructure for sustainable orthopedic MSO platforms built around the principles described above is being built in pockets with the benefit of hard-won operational lessons from the PE era and a technology and policy environment more favorable than at the start of the cycle.

Challenges create opportunities for the willing, but the window will not stay open indefinitely. The definition of physician independence is evolving, but the future offers a little something for everyone.

For distressed PE platforms: The turnaround math is unforgiving but operational improvement is achievable and matters. The highest-return near-term initiatives are closing operational integration gaps and aligning with physicians to prevent attrition. Implant standardization, ASC optimization, and nascent VBC, COE, and AI infrastructure should exist before pursuing a transaction. In a thin buyer market, differentiation is the difference between options and none.

For the surgeon currently inside a distressed PE platform: The waterfall math described in Part 1 does not improve with time. The physician who understands their equity position clearly, models their waterfall honestly, and makes a deliberate decision is in a better position than one who waits for clarity that the capital structure cannot provide. Knowing your worth and your options is paramount.

For the independent surgeons or groups: The infrastructure to compete at scale without surrendering equity or autonomy is coming within reach. Predictable RCM, group purchasing, shared data infrastructure, and AI-assisted administration are accessible at costs that no longer require an institutional balance sheet. The governance discipline required to build an independent MSO that actually works is the harder problem. The frameworks presented in Parts 4 and 5 are a good place to start.

For the surgeon employed by a health system: The distressed PE cycle will surface talent, clinical infrastructure, and in some cases ASC assets over the next 24-36 months that were not available during the consolidation wave. An independent platform that recruits strategically during this period can build in a year what would have taken five years in 2019. If you are considering your options, this is the moment to be actively evaluating them rather than waiting for the market to settle.

For CMS and CMMI: The June 9 comment deadline on the FY 2027 IPPS Proposed Rule is the immediate opportunity. The ASC participation RFI and the physician-owned hospital voluntary participation RFI are both openings for the organized orthopedic community to put a specific, evidence-based alternative framework on the record. Hospitals and health systems will oppose these efforts. That opposition should be weighed against the peer-reviewed evidence on physician group performance, not treated as a reason to avoid the conversation. Meanwhile, the BPCI-X framework is executable and closes a significant gap in current programs.

For implant manufacturers, distributors, and payers: The orthopedic MSO that delivers documented outcomes, direct employer contracting capability, and genuine care coordination is a categorically different partner than a distressed PE platform managing to its debt service. The Geisinger/Medacta relationship is the early expression of what manufacturer-platform alignment can look like when incentives are aligned.

Market conditions made the consolidation of Orthopedic practices inevitable. The PE form it took was a design choice made under specific financial conditions that produced the current outcomes. The next form is also a choice, and the evidence on which design produces better outcomes for patients, physicians, employers, and the broader healthcare system exists.

The platforms that will define the next phase of orthopedic care delivery are being built by people who understand that physician ownership, aligned incentives, and a long time horizon produce something an extractive model never could.

The evidence agrees with them.

Subscribe now

Author Note

I’m an orthopedic surgeon and physician executive who writes about healthcare policy, physician practice economics, and MSK value-based care at The Surgeon’s Record.

I spent years at the intersection of orthopedic practice and healthcare policy — as a surgeon in private practice, an equity holder in a PE-backed MSO, a physician executive at a healthcare services startup, a leader in national MSK organizations, and now as an advisor and advocate.

I’m genuinely curious about what comes next for Orthopedics and welcome conversations with surgeons navigating distressed platforms, leaders building the next generation of physician-owned infrastructure, capital partners interested in an insider’s perspective, and policymakers working on the payment models that will determine what comes next.

I can be reached through The Surgeon’s Record or on LinkedIn.

Questions, thoughts, and comments encouraged.

Ben Schwartz, MD, MBA

Editor-in-Chief

With extended hold periods of PE-backed Orthopedic MSOs, it felt like a good time to take a look at where things stand, particularly from the surgeon’s viewpoint. Many are probably wondering what delayed exits mean for their equity, and what a realistic exit looks like.

Physicians may only have a cursory understanding of what they originally signed — the cash at close, the rollover equity, the expectation of a secondary transaction (“second bite”) at a higher multiple. They may not fully understand where their common equity sits relative to platform debt, the preferred return obligations ahead of it, and the reality of exit multiples given current market conditions.

Publicly available information about capital markets, reimbursement trends, and exit activity is less favorable than original projections suggested for most ZIRP-era platforms. The gap between entry valuation and achievable exit value, after accounting for debt service and preferred return obligations, leaves physician common equity in a narrow or impaired position across much of the cohort.

I am writing this from an informed vantage point. I practiced in a PE-backed orthopedic group during the peak of consolidation, held a senior operating role at a venture-backed MSK company, and am now building a portfolio career in MSK-adjacent consulting work. I no longer hold any position inside a PE platform but do have stock and stock options in the VC-backed company.

This article is not specific to any company, platform, or approach. Rather it represents my analysis of the traditional private equity-backed Orthopedic MSO market based on current trends and conditions. The goal is to use both my current distance and prior proximity for an informed yet detached analysis.

Part 1: The Investment Thesis

This is a mid to late-cycle analysis, not a retrospective. The PE orthopedic consolidation wave may have crested, but the market is far from settled. Many platforms are still operating, many physicians are still inside these structures, and the policy/fiscal environment continues to evolve. The purpose of examining the thesis and its vulnerabilities is not to render a verdict on past decisions but to inform present and future ones. This analysis is for physicians evaluating their options, capital allocators assessing the sector, and policymakers designing payment models that will determine what physician practice structures are economically viable going forward.

With that framing established, let’s review how we got here.

Between 2017 and 2022, private equity consolidated orthopedics at an unprecedented pace. The specialty presented the characteristics PE investors favor: a fragmented market of independent practices, high procedural volume, favorable demographics generating durable demand, robust ancillary revenue opportunities, and a reimbursement structure that historically rewarded high-acuity procedural care. By 2022, more than 16 PE-backed management platforms were actively consolidating orthopedic practices across the country.

The timing was aided by favorable market conditions, both financial and clinical. Near-zero interest rates made leveraged buyouts cheap with far less onerous debt service. The math of rolling up practices at 6-8x EBITDA and exiting a scaled platform at a 12-14x multiple works on paper when cost of capital is negligible. (As we’ll cover later, things get trickier when rates go up and entry and exit multiples invert.)

Perhaps the biggest catalyst was the site-of-service shift to ambulatory surgery centers, particularly for high-acuity procedures historically performed in hospitals. For decades, total joint replacement and complex spine surgery were classified as inpatient-only procedures, meaning Medicare reimbursement required hospital admission. CMS removed total knee arthroplasty from the Inpatient Only List in 2020, followed closely by total hip arthroplasty in 2021.

Whether prescient or opportunistic, the timing was right. The outpatient shift began years ago in the commercial market but crested at the exact moment PE capital was flooding into the space. The pandemic acted as an accelerant as patients and surgeons became comfortable with the ASC setting for higher-acuity cases out of necessity.

From a financial standpoint, ASC economics matter more than professional fees. A total joint replacement generates modest revenue for the surgeon — roughly $1,200-1,500 per procedure. The bigger economic opportunity lies with facility fees. Medicare pays ASCs approximately $9,300 for a joint replacement while commercial rates are often $15-25k+. Migration of joint replacement and spine procedures to surgery centers unlocked facility fees for lucrative Orthopedic procedures creating a massive arbitrage opportunity.

The revenue available at the ASC level was the economic engine the PE-in-Ortho thesis was built around. Robust growth projections for outpatient joint and spine procedures painted a favorable picture and path to grow EBITDA. Platforms that acquired or built ASC infrastructure early captured those economics directly and positioned themselves to take advantage. Platforms that did not were relying on efficiency gains and economies of scale with capped upside.

The physician calculus made sense too, at least pre-pandemic. Independent orthopedic groups were facing relentless administrative burden, rising overhead costs, and deteriorating reimbursement rates. Health systems were aggressive acquirers of independent practices throughout the 2010s, and in many geographies surgeons who had spent careers building independent groups watched hospital systems completely reset market dynamics.

Hospital employment of physicians grew from 23.4% in 2012 to 34.5% in 2024 — a 47% increase over twelve years. In orthopedics, 54% of surgeons remained in private practice as of the 2024 AMA Physician Practice Benchmark Survey, tied with ophthalmology for the highest percentage. That finding reflects the specialty’s historically stronger economic position compared to primary care, driven in part by the ancillary revenue capture that made Ortho practices attractive to PE.

Still, despite their proclivity for independence, Orthopedic Surgeons are not immune to market pressures. When a health system acquires the dominant orthopedic group in a mid-sized market, remaining independent groups lose referral access, imaging revenue share, and payer contracting leverage simultaneously. Health system consolidation created a binary decision: affiliate with a hospital system or seek out an alternative path to maintain independence.

For some, private equity appeared to offer such a path, but the devil was in the details of deal structure.

Deal Structure

Understanding deal mechanics is essential to understanding the current environment. A typical PE transaction worked as follows:

The platform — a PE-backed management services organization (MSO) — acquired a practice at an agreed valuation, typically 6-10x EBITDA for an individual group. The purchase price was paid in two parts: a cash component at close and a stake in the newly formed MSO (rollover equity). Most deals were structured as 60-70% cash and 30-40% rollover equity, enough to ensure physicians had a vested interest in continued platform growth and success.

For senior partners approaching retirement, the cash component functioned as an effective liquidity event at a multiple they could never have achieved selling to a hospital, winding the practice down, or through a buyout. Meanwhile, rollover equity became a potential golden parachute, offering a “second bite of the apple” when the platform exited at a higher multiple in five to seven years.

For younger partners, the value proposition of a private equity sale is less clear. Depending on deal structure, cash at close was often smaller relative to earning years ahead. Furthermore, a smaller rollover equity stake meant less upside in the event of a second transaction. That imbalance was offset by the promise of third or even fourth transactions during the course of a long career, each an opportunity to capture more liquidity at increasing value.

The principal mechanism that funded MSO operations — and created EBITDA for the platform — was the salary scrape. Before the transaction, orthopedic groups typically ran on a net income model: revenue minus expenses, distributed to partners. After the transaction, a portion of that net income was redirected to the MSO as a management fee and to support platform growth and additional acquisitions. Physicians kept their salaries but took home less than they had before with the difference flowing upward to fund the platform’s overhead and debt service, and to generate the EBITDA that would be used to value the platform at exit.

Older partners who negotiated maximum cash at close minimized their exposure to the salary scrape and the downside risk of their rollover equity. In other words, they realized most of their value at the time of the first sale. Younger partners were exposed to income compression with less upfront cash to offset the salary scrape. They were also more exposed to the ultimate performance of the capital stack sitting above their common equity. As a result, early career surgeons will bear the biggest burden of distressed platforms.

The MSO approach made sense for both sides at the time. PE firms followed their playbook: identify a fragmented market with favorable economics (Orthopedics), deploy capital at the available cost (near zero interest), and build platforms that generate value through growth, efficiency, and economies of scale (consolidate). Physicians who transacted did so in response to margin pressure and health system encroachment. They took liquidity and bet on a larger platform to deliver stability and growth capital while allowing for some measure of independence. The prospect of a second liquidity event further sweetened the pot.

Now, slowing sector growth and the paucity of exits hints at the challenges genuine orthopedic consolidation requires. Fund timelines (typically 5-7 years) incentivize rapid EBITDA growth, not years of operational integration, data infrastructure, care model development, and payer relationship building. An investment thesis that made sense during the ZIRP/ASC boom era now looks shakier in retrospect.

Orthopedics: Attractive, Yet Complicated

Orthopedics was not the first specialty to attract PE consolidation. Dermatology, gastroenterology, ophthalmology, and dental service organizations came first. The varying results seen in those specialties provide insights into the orthopedic experience and the limits of generalizing from it.

Ophthalmology and gastroenterology — both ASC-focused, procedurally standardizable, and less dependent on individual surgeon variation for volume — have produced reasonably durable platforms. GI Alliance was acquired by Cardinal Health at a valuation of $3.9 billion, one of the largest specialty PPM exits to date. Of note, Cardinal is a distributor, not a traditional PE firm, suggesting the exit market for specialty platforms may increasingly depend on non-PE strategic buyers. Dermatology has produced mixed results; the aesthetic segment, with higher cash-pay revenue and more standardizable workflows, has performed better than the medical dermatology segment.

On the surface, Orthopedics mirrors Ophthalmology and Gastroenterology with reproducible unit economics, care protocolization, and site of service arbitrage opportunities. On closer analysis, several important differences emerge. First, high revenue musculoskeletal procedures are more complex and tied to individual surgeon relationships, referral patterns, and skill set than most other specialties. Post-transaction productivity and facility utilization incentives are more fragile operationally than anticipated.

Second, procedure-level reimbursement compression continues for the two highest-volume orthopedic procedures, total knee and total hip replacement. These procedures have experienced the sharpest reimbursement declines, directly impairing the revenue base that platforms were underwritten on. Facility fee capture offsets this compression; however, in the typical PE-backed MSO structure, that benefit is appreciated at the platform level, not the surgeon level. That imbalance starts to fray with prolonged hold times and delayed exits.

Because of these factors, PE-backed Orthopedic consolidation hasn’t necessarily been smooth sailing. The fundamental miscalculation was that most attractive acquisition targets already had favorable economics — busy surgeons, profitable ASCs, and established ancillary service lines. These were mature businesses facing headwinds, not distressed assets in need of turnaround.

The ASC piece remains important, but not for the initially intended reasons. ASC distributions have become the main source of salary scrape “repair” but may be insufficient if ownership stake is low or non-existent. In addition, ASC revenue is flowing to debt service obligations caused by stubbornly high interest rates and to address rising overhead costs caused by sticky inflation.

Rather than becoming the main source of EBITDA growth, ASC revenue is being used to keep platforms afloat. Young total joint and spine surgeons understand the value they bring to centers and will be less inclined to let someone else capture that value. Sophisticated groups understood that capturing and holding onto ASC ancillary revenue themselves was the smarter path to independence.

Meanwhile, PE-backed MSOs may be experiencing both buyer and seller’s remorse.

Part 2: A Distressed Cycle

PE-backed consolidation activity has slowed significantly. The deal announcements, press releases about platform expansions, and trade press roundups that characterized the 2019-2022 period have given way to a quieter environment marked by extended hold periods and balance sheet restructurings.

The exit market — or lack thereof — tells the story. PPM recapitalizations hit their lowest level in a decade in 2024, with only 13 completed transactions across all specialties (down from nearly 100 in both 2021 and 2022). It’s worth noting that a rebound in M&A and exit activity has been forecast by industry analysts in each of the past three years.

The most recent iteration of this forecast, predicting that PE will "reignite" PPM M&A in 2026, follows the same pattern of optimism generated by advisors and bankers hopeful for transaction fees. This year could be the year but hope for continued interest rate cuts is fading in the face of domestic and global economic uncertainty. Inflation remains high, and there are now whispers of rate increases. The spread between buyers and sellers persists, portending less activity and/or lower exit multiples. Despite best efforts, willing an exit market into existence has not proven successful.

In Orthopedics, the most-cited exit was SCA Health’s acquisition of OrthoAlliance from Revelstoke Capital Partners, widely presented as validation of the second-bite model. But that exit might be a difficult-to-replicate exception, not evidence of a broader thaw. Orthopedic PPMs present a management challenge that ASC operators like SCA Health may have underestimated. The value of the platform is inseparable from that of the physicians, who retain leverage, geographic mobility, and the ability to affiliate with competing facilities. Managing a collection of independent-minded orthopedic surgeons across dozens of markets is different than operating a standardized ASC chain.

Meanwhile, SCA itself has since undergone significant organizational restructuring, limiting its utility as a template for subsequent transactions. “Just sell to Optum” is no longer a solid exit plan. A clear buyer’s market for Orthopedic MSOs has yet to materialize while the seller’s market is being pressured by long hold periods. In turn, those hold periods and the financing structures associated with them are creating another layer of vulnerability.

Most 2019-2022 vintage orthopedic MSO platforms were financed with floating-rate debt from private credit lenders. The Secured Overnight Financing Rate (SOFR), used as a benchmark reference rate for most institutional lending, rose from near zero in early 2022 to over 5% by mid-2023. A platform carrying $300 million in debt at SOFR plus 600 basis points saw its effective borrowing rate go from roughly 7% to over 11%, representing an additional $12-15 million in annual interest burden on the same principal. Growing debt service exacerbated by prolonged hold times directly impairs EBITDA at a time when compressed valuation multiples make it even harder to achieve a favorable exit. As a result, some platforms turned to continuation funds.

Orthopedic Care Partners, one of the larger MSOs backed by Varsity Healthcare Partners, completed a $543 million recapitalization in late 2024 — a $185 million hybrid capital raise from Brookfield Asset Management alongside a $358 million senior credit facility refinancing led by TPG Twin Brook Capital Partners. While framed as positioning for growth, from a different lens it could be interpreted as a balance sheet extension. By refinancing maturing debt and adding a capital cushion, OCP could be buying time, hoping a stronger exit market materializes. The platform simultaneously installed new executive leadership, perhaps repositioning for a transaction process.

The distress in the current cohort is only partly a function of bad luck with rates and bad timing with macroeconomic conditions. There’s a fundamental flaw in the highly leveraged PE model as applied to orthopedics. Short fund timelines create little incentive to do the hard, slow work of genuine operational consolidation. Building centralized RCM infrastructure, standardizing supply chain, integrating cross-platform benefits, developing payer contracting leverage, and embracing value-based care capability all take, at minimum, 3-5 years to generate returns. A PE fund on a 5–7-year hold cycle, already 18 months into platform formation when these initiatives would need to begin, has limited incentive to pursue them. The playbook instead is to affiliate practices quickly and generate volume-driven growth before underlying economics become regressive.

The lack of M&A and exit activity may be an indication that the latter has already happened.

Understanding Capital Structure

The result of prolonged hold times is a cohort of platforms caught between an entry valuation they can’t recover, a debt service burden compressing the EBITDA they need to attract a buyer, and a physician workforce that is increasingly aware that the second bite may not materialize. If an exit does happen, physicians are often at the mercy of capital structures and waterfall mechanics that were opaque at the time of the initial sale. These mechanics are particularly salient when valuations and exit multiples turn unfavorable.

Consider a theoretical platform that enters with $20M EBITDA at a 12x multiple, or $240M in enterprise value. Capital is structured as $156M in senior floating-rate debt and $84M in equity split 70/30 between PE preferred and physician common equity. Effective debt rate at entry is approximately 6% (SOFR near zero plus spread), and the annual interest burden is approximately $9.4M.

The original underwriting assumption is that EBITDA grows to $26M over the hold period (30% growth through rollups, operational improvements, and volume growth) and that an 11x exit will occur by year 5 at an enterprise value of $286M. At that exit, debt is repaid ($156M), PE preferred equity receives $88M (1.5x return on $59M), and physician common equity receives $42M — a meaningful return on the rollover. This is the enticing “second bite” realized. It also represents a best-case scenario that hasn’t broadly materialized.

First, SOFR rose from near zero to over 5% by mid-2023. Effective debt rate moved from ~6% to ~11.5%. In our scenario, annual interest burden would increase from $9.4M to approximately $18M, an $8.6M annual drag on free cash flow that was supposed to fund operational investment and EBITDA growth. Rather than pumping up EBITDA, ASC facilities are being diverted to growing interest payments.

Second, EBITDA growth underperformed. Reimbursement compression, misaligned physician incentives, and limited operational integration produced modest growth — let’s say $22M rather than $26M, a 10% increase rather than the budgeted 30%. In cases of PIK (payment-in-kind) interest provisions, unpaid interest is added to principal rather than paid in cash, causing the debt balance to grow from $156M to $170M or beyond.

Third, exit multiples contracted. Sponsor-to-sponsor recapitalizations, one expected exit mechanism, effectively ceased for most platforms in 2023-2024. Current market multiples for orthopedic platforms without demonstrated operational EBITDA growth are likely in the 7-9x range (or less) rather than the 10-12x range assumed at entry.

That presumed favorable exit at the time of the initial sale now looks like 8x on $22M EBITDA or an $176M enterprise value. After repaying $170M in senior debt, only $6M remains. All of it goes to PE preferred equity; physician common equity goes to zero. Even in a moderate scenario — 9x on $22M = $198M — senior debt consumes $170M, leaving $28M against $88M in PE preferred obligations. Physician common equity still receives nothing. In our scenario, given current exit multiples of 7-9x, EBITDA would have to grow 44-65% before physicians see any return at all.

The specific dollar amounts here are theoretical and illustrative. However, the general waterfall mechanics hold. In any ZIRP-vintage transaction with 60-65% leverage and standard preferred return terms, physician common equity requires an exit value exceeding the original entry valuation just to break even — a threshold that multiple compression and modest EBITDA growth have placed out of reach for much of the cohort. And that’s not even factoring in transaction fees and other hidden costs that further reduce distributable funds.

These compressed scenarios do not require catastrophic operational failure. Instead, they require the simultaneous occurrence of three conditions that were each individually unlikely to be modeled at signing: higher interest rates, modest EBITDA growth, and compressed exit multiples.

All three exist now. Not every platform sits at this exact intersection, but many fall within a band where even moderate underperformance produces similar waterfall compression.

Is Recovery Possible?

There exist conditions under which the PE orthopedic model produces acceptable outcomes. In other words, the model is not inherently unworkable. However, it is sensitive to conditions that turned unfavorable between 2022 and 2025. For ZIRP-era platforms, the model produces acceptable outcomes for common equity holders (physicians) under four conditions:

Condition #1: EBITDA growth sufficient to justify entry multiples.

Most platforms were acquired at 10-14x EBITDA. For common equity to recover meaningful value at current exit multiples of 7-9x, a platform needs to grow EBITDA materially above the entry base. The principal levers to achieve this growth include volume expansion, ancillary revenue capture (e.g., ASC facility fees), operational savings, and improved RCM. The platforms that invested in operational integration early and can demonstrate this growth trajectory are in a better position than those that relied primarily on practice affiliations and arbitrary volume targets.

Condition #2: A viable balance sheet extension mechanism that buys time without deepening the equity trap.

When a clean exit is not available, PE funds have two related, yet distinct options for extending the hold period: continuation funds and hybrid recapitalizations. A continuation fund transfers the asset from an aging fund into a new vehicle managed by the same GP. Existing LPs can cash out or roll equity into the new vehicle; the GP retains control, and the hold period resets without a change of ownership. Physician owners may experience some liquidity but may also be subject to new, possibly less favorable, contract terms.

A hybrid recapitalization, like the OCP Brookfield/TPG Twin Brook transaction, brings in new third-party capital at the platform level, refinancing maturing debt, adding equity cushion, and extending maturity while the existing GP retains its position. Both mechanisms can be useful for platforms with genuine underlying EBITDA momentum that simply need more time. However, neither changes capital structure math nor waterfall provisions.

Condition #3: An exit pathway at a valuation sufficient to satisfy obligations ahead of physician common equity.

A true exit requires a buyer — either a new PE sponsor acquiring a controlling stake in a sponsor-to-sponsor recapitalization, or a strategic acquirer such as a health system, distributor, or payer. Sponsor-to-sponsor recapitalizations were the expected primary exit mechanism for most 2021-vintage platforms. As detailed in the exit options section below, that market has been largely closed since 2023. Strategic acquirers — principally health systems — represent the most plausible exit pathway for the distressed cohort. But their acquisition terms may fall short of physician expectations set at the time of the first sale.

The conditions under which a true exit generates meaningful physician common equity recovery are the same as Condition #1: exit proceeds must clear the combined debt and preferred return threshold which requires either a strong operational EBITDA trajectory or a recovery in ZIRP-era exit multiples. For most of the pandemic-era cohort, neither condition is currently met.

Condition #4: Physician stability must hold through the resolution period.

Platform EBITDA and a convincing growth story depend on physician equity holder buy-in. Surgeon volume depends on satisfaction with compensation, governance, and the clinical environment. A platform experiencing income compression and governance dissatisfaction faces a compounding dynamic: attrition —> reduced EBITDA —> increased financial pressure —> more attrition. Physician retention is simultaneously the leading indicator of platform health and the variable most difficult to manage in a distressed capital structure. Moreover, prolonged hold period breed discontent. The sunk cost fallacy only holds for senior surgeons who experienced a favorable first sale. Younger surgeons are less incentivized to wait out an exit.

Conditions #1 and #4 might be achievable for a subset of platforms while Conditions #2 and #3 are largely subject to market whims. The practical implication is that prolonged hold times and exit pressures are in direct conflict with unfavorable market conditions.

Something has to give.

Possible Outcomes

The capital mechanics described above produce a range of outcomes depending on the rate environment, exit market, and platform operational performance. There are four plausible outcomes for the low-interest rate cohort that’s currently pushing the 5–7-year timeline.

Scenario #1: Favorable Outcome

Conditions: SOFR declines to 3-3.5%, platforms demonstrate 15+% EBITDA growth above their entry bases, PE deal activity recovers meaningfully
Outcome: Sponsor-to-sponsor recapitalizations resume for strong platforms, strategic buyers provide exits for mid-tier assets, and common equity holders in the better-structured platforms realize meaningful value (1.5-2x return on rollover equity).

Scenario #2: Acceptable Outcome

Conditions: Rates stabilize at current levels, platforms demonstrate 5-10% EBITDA growth, limited sponsor-to-sponsor activity resumes
Outcome: A subset of platforms find exits over 2026-2029 at reduced but acceptable multiples (0.5-1.5x return on rollover equity).

Scenario #3: Distressed Outcome

Conditions: Rates stabilize or increase, EBITDA growth is flat, the buyer’s market remains weak
Outcome: Distressed sale to a health system, structured refinancing, or a wind-down, with physician common equity recovering little to nothing (rollover equity goes to zero).

Scenario #4: Catastrophic Outcome

Conditions: Rates increase, physician attrition accelerates, loan covenant breaches trigger, acquirers remain constrained by their own capital challenges
Outcome: The distressed resolution timeline accelerates and the buyer price at which assets clear is insufficient to satisfy preferred return obligations for many platforms. Not only is common equity wiped out, but debt restructuring or loan forgiveness triggers Cancellation of Debt Income, a taxable event for physicians.

Part 3: Exit Options

Capital sponsors and physician common equity holders are both hopeful of a return to exit activity. In the current environment, there are four principal exit options, each with its own constraints that explain the current backlog.

Option #1: Recapitalization and continuation vehicles

The most common mechanism for extending PE holds beyond fund term limits is the continuation vehicle. To review, this fund structure allows GPs to transfer assets from an aging fund into a new vehicle, providing partial liquidity to existing LPs while extending the hold period for investors who choose to roll into the new vehicle. This has been used in the orthopedic PE cohort, but there are limitations.

First, continuation vehicles do not solve the valuation mismatch problem. Instead, they satisfy LPs’ desire for liquidity and extend the hold period until platform financials and/or exit conditions improve. Furthermore, the asset is transferred at a negotiated valuation that must satisfy both rolling and exiting LPs. With compressed multiples, that “spread” may be difficult to resolve to the satisfaction of new and existing investors.

Second, the rate environment makes the new vehicle’s economics less attractive than the original fund’s underwriting assumed. A continuation vehicle financing its acquisition of the asset at current debt costs starts from a higher cost basis than the original platform leaving less room for the return that would attract rolling LPs.

Finally, physician equity holders may find their rollover position restructured or diluted in the continuation vehicle transaction depending on how the new vehicle is capitalized. The continuation vehicle extends the timeline but does not change the underlying truth — exit value must exceed obligations senior to physician common equity. That gap has not closed.

Option #2: Sale to a health system

Health systems are the most logical strategic acquirer for distressed orthopedic platforms. Constraints on this pathway are related to physician expectations and the realities of health system acquisitions.

Health systems do not want to acquire loan obligations. A platform carrying $150-300+ million in senior debt is not an attractive acquisition target for a health system whose own balance sheet is under pressure and whose credit rating is sensitive to leverage ratios. Most health system acquisitions of physician practices are structured as asset purchases, not enterprise acquisitions that include existing debt. That means the PE fund and its lenders need to resolve capital obligations before or as part of the health system transaction. This typically requires the sale price to cover the debt and preferred return obligations before physician common equity sees any value — essentially a wipeout scenario.

Health systems also do not pay strategic multiples for orthopedic practice volume. What they want is surgeons who commit to using the health system’s facilities, cases that flow through their OR and HOPD, and facility fee revenue from ASCs they can acquire and control. The alignment between what health systems want to buy and what distressed PE platforms are positioned to sell is imperfect.

Further complicating matters, platform physicians may resist becoming employees of a health system, especially if such an acquisition renders their rollover equity worthless. While health system employment provides a stable salary floor, the income ceiling is often lower than traditional private practice. Employed physicians rarely benefit from ancillary income generation, a non-starter for surgeons used to getting distributions from physical therapy, DME, imaging, and ASC revenue. Finally, many groups partnered with PE specifically to avoid health system employment.

Option #3: Sponsor-to-sponsor sale

A larger fund acquiring a distressed orthopedic platform at today’s debt costs needs to underwrite a return on a higher effective cost of capital than the original fund assumed. The business model risk — whether orthopedic PPMs can generate the EBITDA growth required to service debt and produce the return a fund needs — has not been proven at scale. Lack of sponsor-to-sponsor sales despite prolonged hold periods serves as proof. Despite persistent rumors of multiple platforms coming to market, only one true second sale has taken place (the aforementioned OrthoAlliance/SCA transaction).

The operational thesis was that genuine consolidation of back-office functions, payer contracting, and care delivery would compound EBITDA beyond what organic volume growth alone produces. The evidence that this has been achieved systemically is limited. A new fund acquiring a platform is essentially making a fresh bet on operational value creation in an asset where the original fund’s hold period was not long enough to fully test that thesis. The remaining consolidation opportunity — the whitespace of large independent orthopedic groups in attractive markets willing to transact — has contracted significantly from the initial consolidation period, limiting the add-on acquisition growth strategy that supported many original investment theses. For a larger fund to make this bet at current debt costs, the asset needs to be demonstrably well-operated, have genuine ASC economics, and be priced to reflect the current environment rather than the original entry multiple.

Only a small subset of the cohort may be able to satisfy those conditions.

Option #4: Physician buybacks

The most interesting, and least discussed, exit option is a partial or full repurchase of the platform by its physician partners. This pathway deserves more attention than it usually receives because it is the option most aligned with the overall MSO thesis. It’s not an easy or straightforward path, but it might be the most viable.

A physician buyback requires unwinding the existing capital structure. Debt must be satisfied or negotiated down before the practice is recapitalized under physician ownership. In a scenario where the platform’s enterprise value is below the sum of its obligations, this requires either a discounted payoff negotiated with lenders and the PE fund or a transaction in which the physicians acquire the operating assets rather than the platform itself. In this scenario, the PE fund must resolve the financial structure.

Physician buybacks face several obstacles, cash requirements chief among them. An orthopedic group repurchasing its infrastructure needs working capital to fund the transition period, invest in the back-office systems it previously relied on the MSO to provide, and buy out ASC equity that the platform holds. That requirement — typically several million dollars for a mid-sized group — requires either physician equity contributions, bank financing secured against practice revenue, or a third-party capital partner who is not a traditional PE fund. Community development financial institutions, family offices, and specialty-focused credit funds are potential sources; none is as efficient as a PE fund for this purpose.

Physician risk tolerance is the second constraint. Surgeons who transacted into a leveraged buyout that failed to meet expectations may not be enthusiastic about the prospect of a leveraged buyback. The physicians most likely to pursue a buyback are those with the financial reserves to absorb the transition cost and the governance discipline to manage a physician-owned entity. Recreating what existed before the PE transaction is not a given.

Non-compete clauses and restrictive covenants create additional complexity. Transaction documents typically include covenants that restrict physician mobility and define the conditions under which they can separate from the platform. Those provisions were designed to protect the platform’s value during the PE hold period; in a buyback scenario they may constrain the terms on which physicians can repurchase or reconstitute their practice. Each platform’s transaction documents are different, and any buyback requires careful legal analysis of these provisions before structuring begins.

Despite these obstacles, the physician buyback outcome is worth consideration. A successful repurchase converts a failing PE-backed platform into a physician-owned model that evidence suggests produces better cost and quality outcomes. It may also benefit from operational efficiencies built during the PE hold period. Finally, it creates the physician ownership base from which sustainable alternative models can be developed. The obstacles are real but not insurmountable, and in some markets, they will be overcome.

The private equity orthopedic consolidation wave was not built on bad intentions or irrational assumptions. The thesis made sense under the conditions in which it was conceived: a fragmented specialty with genuine consolidation economics, favorable reimbursement trends, a compelling ASC opportunity, and cheap capital. Many of the platforms built during this period created real infrastructure — centralized operations, shared services, ASC networks — that has genuine value independent of what the capital stack above it looks like.

What the thesis did not adequately account for was the interaction between leverage, fund timelines, and an operating environment that never materialized. Higher rates, compressed multiples, and modest EBITDA growth arrived simultaneously, and the waterfall mechanics that seemed theoretical at signing have become very concrete for physician equity holders.

The exit landscape doesn’t offer easy resolution. Continuation vehicles extend the timeline without changing the math. Health system acquisitions rarely generate the proceeds or independence physicians expected. Sponsor-to-sponsor recapitalizations remain largely inactive for the current cohort. Physician buybacks are possible but require capital, governance discipline, and risk tolerance beyond the reach of many groups.

Something has to resolve. The question is on whose terms and whether the infrastructure built during the PE era becomes the foundation for a more durable ownership model or simply a balance sheet problem to be unwound.

That question is the subject of a future piece.

Subscribe now

Ben Schwartz, MD, MBA

Editor-in-Chief

Is this progress or an expensive administrative solution to a problem largely of CMS’s own making?

Is There Still Meat on the Bone?

Original CJR ran for eight years and generated decent savings — roughly $112 million over two years in its later performance years. As anyone who has studied or participated in these models knows, the majority of those savings came from reductions in post-acute care. Sending patients home instead of to skilled nursing facilities generates the biggest financial wins. Better surgical outcomes and lower complication rates are well and good, but discharge destination is the value champion.

Studies show that VBC and FFS payment models deliver more or less equivalent quality. For example, a Health Affairs analysis found that “changes in quality of care, measured by complication rates, emergency department visits, and mortality, were not associated with bundled payment participation.” Programs like CJR reduce costs, but don’t necessarily improve outcomes. As I’ve already pointed out, those reduced costs come from reduced post-acute care spending. Once that’s been exhausted, outcomes driven cost savings may not pick up the slack.

CJR-X is a tweaked version of a similar model, applied nationwide. The post-acute care reduction ship has already sailed in many places. Fewer patients are being discharged to rehabs and SNFs. There’s another factor at play too, site-of-service shift. CMS’ own data shows that, by the end of the original CJR model, 70% of elective joint replacements were being performed in the outpatient setting. By definition, those patients go home. (Note: CJR-X expands to total joints performed in HOPDs. Impact TBD.)

If most savings in the original model came from avoiding SNFs and rehabs, where will the savings come from in the new model? Are we reaching the point of diminishing returns? Original CJR saved $1,142 per episode. CMS projects CJR-X will save $725M over a 5-year period. With around 750,000 Medicare joint replacements performed annually, that’s less than $200 savings per episode — a significant reduction. Notably, cost reduction isn’t even an explicit goal of CJR-X.

The tradeoffs are scale and competition. Everyone who’s not part of TEAM will be forced into CJR-X. Benchmark comparisons are regional, not internal. Program goals may be shifting away from costs savings towards care coordination and quality improvements. These are noble goals, but to what end? Is all the added complexity justified, or are there better ways to achieve the same results?

The Wound CMS Inflicted on Itself

Between 2000 and 2024, inflation-adjusted Medicare reimbursement for primary hip and knee replacement fell 55%. CMS currently pays approximately $1,160 for a joint replacement, including 90 days of postoperative care. In real terms, that’s less than it paid twenty years ago for procedures whose outcomes and efficiencies have improved meaningfully in the setting of an aging, less healthy population.

Reducing unit price in the face of inflation-driven overhead increases pushes rational actors to increase volume. Annual Medicare joint replacement surgeries doubled from 2000 to 2019. Over that same time period, reimbursements experienced negative CAGRs. If the contention is that FFS incentivizes volume, no one should be surprised when lower payments lead to more procedures.

Of course, that assumes that “volume over value” is a valid argument. The easy conclusion is that lower reimbursement produces unnecessary procedures — surgeons stretch indications to compensate for smaller margins. The evidence to support that assumption isn’t as strong as we think.

Joint replacements have pretty well-defined criteria with fewer gray areas than, say, back surgeries. There will always be quick-to-cut surgeons, but overutilization of joint replacements remains an open question. A controversial 2015 study deemed 34% of knee replacements “inappropriate.” That study had significant methodological flaws and was based on outdated appropriateness criteria. In some populations, joint replacements are arguably underutilized.

Volume growth is real, but shifting demographics, not surgical opportunism, is the main driver. What we have is an aging population that’s being diagnosed with hip and knee arthritis at increasing rates. There is absolutely no need to do unnecessary joint replacements — surgeons will need to double their caseload by 2050 just to meet expected demand. Tinkering with payment models won’t change this dynamic.

No one has a great answer for the demand side problem. Mandatory bundles don’t reduce the number of people who need a new hip or knee. Episode-based accountability doesn’t slow the demographic and metabolic trends driving volume. The supply of fellowship-trained arthroplasty surgeons is not infinitely elastic. Current reimbursement trends may make the subspecialty less attractive to trainees and force more surgeons to opt out of Medicare. An access problem is bubbling underneath the cost control conversation.

Compressed reimbursement also makes independent practice less economically viable. Consolidation of independent groups is demonstrably inflationary. Once surgeons are employed, market power increases, facility fees tacked on, and the balance of power shifts. As a price setter, Medicare is somewhat protected from consolidation forces. But vertical integration and site-of-service payment differentials undoubtedly increase costs for the government, too. Mandatory VBC may be the proposed solution, but at least one study found that physician-led bundles save more money than hospital-led bundles. CMS appears to be pivoting away from surgeon-led VBC models.

Fee compression drives increased volume and consolidation. Volume and consolidation are inflationary, and inflation justifies cost controls. VBC models generate modest net savings but come with significant administrative cost. Meanwhile, the underlying structural forces compound and shifting demographics drive utilization. CMS has been treating a condition it created and a demand problem it can’t control with a remedy that could make the condition worse.

A Wrinkle in TEAM

The FY2027 IPPS proposed rule containing CJR-X includes an interesting wrinkle: an RFI asking whether grandfathered physician-owned hospitals should be allowed to participate in TEAM. Unfortunately, it’s not a full reopening of the ACA’s Section 6001 moratorium on POHs — but it’s something.

What the RFI would do is generate controlled, comparative data on the two arguments the AHA has used for years to justify the ban. First, that physician-owned hospitals cherry-pick healthier patients. Second, that their apparent cost efficiency is an artifact of that selection rather than genuine care delivery improvement.

Recent analyses across roughly 740,000 beneficiaries found no evidence of cherry-picking. Commercial prices were 17% lower with potential annual Medicare savings estimated at $1.1 billion. As expected, the AHA pushed back on those findings which were commissioned by physician advocacy groups. TEAM participation by POHs could settle the argument once and for all. If both sides truly believe in their stance, they should welcome a narrowly scoped opportunity to prove themselves right. So far, that’s not what’s happening.

BCBSA has already submitted a letter to the DOJ asking for limited Section 6001 exceptions. They explicitly want surgical-specialty facilities, including orthopedic, spine, and cardiac hospitals, carved out. These are the exact facilities that would generate the most meaningful comparison data. That a commercial payer commented on a proposal 11 months before it was announced is telling and, frankly, more of the same game most are tired of playing.

The AHA and Federation of American Hospitals are expected to file on the same side before the June 9 comment deadline. If these efforts succeed, CMS will have passed up the one provision in its own rule that pointed toward a structural answer rather than an administrative one. It will have once again capitulated to self-protectionism at the expense of meaningful progress.

A Different Test

One way or another, I keep coming back to same conclusion: CMS is making things unnecessarily complicated. The ROI on these models just doesn’t seem to justify their complexity, especially if more effective alternatives are possible. I’ve written before about what a genuinely different model might conceptually look like — The Simple Model. The core argument is straightforward: we keep testing variations of the same complicated hypothesis when the simpler one hasn’t been tested.

A targeted, voluntary test of reimbursement restoration for independent orthopedic practices — with episode cost and quality tracked through existing registry infrastructure and claims data — would test the second-order case directly. CMS has the data, does it really need convoluted, administratively burdensome reporting mechanisms?

It’s a simpler model than CJR-X without the need for 230-pages of proposed rules or administrative infrastructure implementation at thousands of hospitals. The net effect could be slowing or reversal of the consolidation trend and volume migration to lower-cost settings. The net Medicare math is likely favorable. If not, the cost of the experiment is bounded and you’ve learned something.

CMS knows how to adjust conversion factors and construct complicated reimbursement models. Based on the results so far, it may be going about things the wrong way. The same proposed rule that would mandate episode accountability at every hospital in America contains a question about whether POHs should be allowed to compete. It’s a small but extremely important provision. The salve for CMS’ self-inflicted wounds has been visible the whole time.

Primum non nocere.

Subscribe now

Ben Schwartz, MD, MBA

Editor-in-Chief

I never considered dropping Medicare. Like most specialists, I dutifully played by CMS’ rules — even when they seemed unfair, arbitrary, or self-defeating. When the treadmill sped up, I ran faster. When the rules changed, I adjusted.

Then again, what choice did I have? It’s easier to leave the Mafia than it is to opt out of Medicare. CMS knows this. It has repeatedly called the “opt out” bluff, and docs fold every time.

Joint replacements are among the most successful interventions in the history of medicine. They are life altering for patients and provide tremendous societal benefit — up to $30,000 per patient. To match that level of value, the price of GLP-1s would have to drop 80%. Modern joint replacements last a lifetime. GLP-1s have to be taken forever.

Despite their already high value proposition, CMS is constantly squeezing the margins on joint replacements. Declining reimbursements are one thing. But every newly announced CMMI innovation program seems specifically designed to deemphasize the role of specialists.

Does CMS hate specialists?

Programs like ACCESS, TEAM, ACO LEAD, and the recently unveiled CJR-X consistently treat specialists like followers, not leaders. Why does CMS seem determined to route care around specialist expertise rather than through it? Is this circuitousness the best way to achieve accountability? Do gatekeeping and forced collaboration trade short-term value creation for long-term value erosion?

CMS may or may not hate specialists, but I’d argue it’s grossly misinterpreting their value.

Payment Reality Refresher

I covered the payment picture reality in detail last week, but here’s a brief refresher. There’s broad sentiment that specialists pay is too high and primary care pay is too low. Many, including the head of CMS, believe that dynamic needs to change. The truth is, it already has.

The short version is that inflation-adjusted Medicare reimbursement for joint replacements has declined 57% over the last 20 years. Other surgical specialties have experienced similar, though less severe, declines (Figure 1). The 2026 Medicare Fee Schedule included an across the board 2.5% efficiency cut for almost all procedures.

You could argue that CMS is effectively "taxing" the perceived efficiency of surgeons to subsidize the "cognitive time" of primary care. The point here is not to fall into the unhelpful specialists v. PCPs compensation debate. That’s a relic of budget neutrality, not sophisticated, innovative care models. As an aside, I’d argue the two should be decoupled entirely.

The more important point is that perpetuating the tug-of-war between primary and specialty care is probably counterproductive to CMS’ goals. You don’t foster collaboration by putting your thumb on the scales. Nor do you derive value by creating an accountability trap that catches patients in the middle.

The Accountability Trap

CMS wants all beneficiaries in some form of accountable care by 2030 — a goal that has persisted across administrations. Since its inception 20 years ago, CMMI has spent significant time and resources building accountable care architecture. Despite an uneven start, the numbers are starting to support the thesis. Moving away from fee-for-service towards outcome-based accountability feels right, no argument there. But the structure seems off.

The Logic

The premise of the argument goes something like this:

1. Specialty care is expensive

2. Avoiding specialty care decreases costs

3. The best way to avoid specialty care is to shift accountability (i.e., control) away from specialists

Corollary: Taking accountability away from specialists forces collaboration

This thesis is reflected throughout recent CMMI models. For example, ACO LEAD is primary care-led by design. Specialists participate as Preferred Providers through CARA which turns ACOs into conveners (with CMS’ help). Meanwhile, specialists take on episode-based financial risk with less data access and less organizational authority than the ACO creating their benchmarks. ACOs might be accountable for the total cost of care, but what happens when their upstream decisions skew downstream results?

LEAD isn’t the only program that attempts to preempt specialists. ACCESS addresses chronic MSK conditions but routes intervention through technology platforms. TEAM and CJR- X place accountability at the health system level. The common thread in all these programs is the lack of explicit specialist leadership.

This subversion of specialty leadership contradicts the evidence. A JAMA analysis of 91 orthopedic groups in BPCI Model 2 found reduced Medicare payments and improved clinical outcomes with no change in patient complexity. Savings came from reduced utilization of postacute services (SNF discharge rates, home health utilization, readmissions) — postop care decisions best driven by specialists.

Another study found that physician group practices saved $240 more per BPCI-A episode than hospitals. Surgeons were also more likely to participate even though the model was voluntary. The authors presciently concluded that “future models should consider strategies to align physicians when participation is restricted to hospitals.” That restriction is occurring now; the alignment isn’t.

If specialists perform well when held accountable why cut them out of the loop?

Who Watches the Watchers?

The working assumption throughout these models is that shifting accountability away from specialists will lower costs and drive better decisions. But data are only as useful as the person interpreting it. If your LEAD dashboard shows a 20% SNF discharge rate post joint replacement, should it be up to a primary care ACO to determine whether that's appropriate and acceptable?

Who can distinguish a risk stratification problem from a surgical quality problem from a post-acute management problem? Who identifies which surgeons are cherry-picking healthy patients and which are managing a genuinely complex population? Who manages edge cases that don’t fit neatly into defined appropriateness criteria?

Does such a person exist in primary care-led ACOs?

Don’t get me wrong. There are certainly situations in which MSK conditions can be appropriately evaluated and managed by non-specialists. But how do you make that determination and who’s best to make it? What’s best for the patient? We universally agree that non-specialist medical directors working for payers shouldn’t be making specialty care UM decisions. Why accept that same mechanism inside a VBC program?

Payment model doesn’t change accountability or clinical rigor, nor should it.

[Sidebar: AI is a possible solution to close this gap, but using AI to upskill PCPs doesn’t solve the fundamental problem. AI or not, do PCPs want to be accountable for specialty level care? Should specialists be accountable for the consequences of those decisions?]

Collaboration: Theory v. Reality

CMS seems to believe that collaboration is the answer to the accountability problem. Collaboration sounds great in theory. In an ideal world, PCPs and specialists would closely co-manage patients to reduce fragmentation and gaps in care. But collaboration isn’t what it used to be.

PCP and specialist collaboration has degraded for a variety of reasons — employment and consolidation of PCP groups, siloing of care, and EMR-driven inefficiencies to name a few. Tech can foster collaboration, but inbox fatigue and the “garbage in, garbage out” remain. Frontline reality is that many physicians, specialists and PCPs alike, simply don’t have the time or bandwidth for meaningful, consistent collaboration. Years of self-inflicted structural and systemic wounds led to this point. Payment models won’t lead us out.

Unfortunately, collaboration simply doesn’t happen in the real world to the level it should. Incentivizing collaboration through payment model design risks making it another gameable metric that sounds good in theory but ends up perfunctory in practice. Ideally, you’d fix the underlying structural problems that made collaboration so difficult in the first place.

Lessons from BPCI

CMS seems to believe that accountability works best when shouldered by PCPs. What if the inverse were equally true? What if specialists also take accountability for whole-person care? What happens when specialty care becomes the entry point for prevention and chronic care management?

What if we already had the answer?

Staying in Your Lane

BPCI taught us important lessons in accountability. Orthopedic Surgeons quickly learned that improving a patient’s overall health was a critical component of bundle success. BPCI made it no longer acceptable to cut first and worry about the consequences later. Specialists now had to consider the entire patient and the impact of social drivers, mental health, and other holistic factors.

That didn’t mean Ortho Surgeons started managing blood sugars or titrating hypertension meds. Instead, many hired nurse navigators, established preoperative optimization clinics, and developed surgical appropriateness guidelines. Rather than gatekeepers, surgeons became traffic cops, identifying modifiable risk factors then guiding patients to the proper experts.

Orthopedic Surgeons didn’t try to route patients away from primary care. In fact, they did the opposite — identifying undiagnosed or undertreated medical conditions and making sure patients followed up with the people most qualified to address them. No one, including the surgeons themselves, thought Orthopedists managing chronic medical conditions was the best thing for patients.

Shouldn’t the reverse be true for PCPs and chronic MSK conditions?

Inversion

Another lesson of BPCI was that MSK conditions are often the canary in the coal mine of declining overall health.

High blood pressure, poorly controlled blood sugars, and early-stage renal failure don’t necessarily drive people to seek medical care. Painful backs and joints do. Specialty care as a gateway to primary care is a hidden inversion that can be leveraged in powerful ways.

A patient with bone-on-bone knee arthritis who wants their life back will lose weight, stop smoking, and get their diabetes under control if you make it a condition of safer, lower risk surgery. Though optimization isn’t a panacea, research shows that some lifestyle changes and health benefits persist.

Patients who manage to stop smoking before surgery are more likely to remain smoke-free after surgery. Similar benefits are seen with perioperative glycemic control. Persistent weight loss is trickier; most patients either stay the same or gain weight after surgery. However, studies specific to weight loss as part of an optimization program are lacking. Still, favorable non-MSK disease management outcomes can be achieved through the leverage point of an elective procedure.

This inversion should be harnessed, not ignored.

Thinking Again

CMS is spending enormous resources on chronic disease prevention programs to move needles that well-designed preoperative optimization protocols already move. Specialists aren’t the problem in the cost equation. In the right model, they’re one of the most underutilized chronic disease management assets in the Medicare system. Dell Medical School’s Musculoskeletal Institute and the Joint Health Program at Duke Margolis are examples of specialist-led whole health can be.

The question isn’t whether PCPs or AI can manage some MSK conditions in some patients. They probably can. The question is who’s best positioned to make that determination and who should be accountable for the consequences when they get it wrong.

CMS keeps building accountability mechanisms that separate financial risk from clinical authority. Programs like ACCESS, ACO LEAD, CJR-X, and TEAM let someone else control the levers while specialists absorb the downside. They limit specialist leadership opportunities while relying on the false premise that specialty care avoidance equates to value.

That theory might be right. It could also be a costly mistake.

Next week in Part Two: a way out of the accountability trap, why CMS’ current approach isn't quite right, and the Ambulatory Specialty Model.

Ben Schwartz, MD, MBA

Editor-in-Chief

To be honest, I’m not much for politics. As a pragmatist, I find the entire political process anathema. There’s a lot of lip service to healthcare reform, but little meaningful action. That may be changing — a little. Anger continues to grow, public figures like Mark Cuban are increasingly critical, and the current administration is taking a non-traditional approach. (For better or worse. Or, maybe for better and worse).

I’ve always felt the most effective way to fix difficult problems is to develop external solutions. This is especially true in healthcare where inertia, incumbency, resistance to change, and perverse incentives create a frustrating, impenetrable morass. No point in waiting around for help that probably isn’t coming.

Perhaps that’s a cynical take, but one that matches ground truth reality for physicians. The regulatory environment has not been particularly kind to us over the last few decades. The net-net of policy change has been greater administrative burden, economic compression, complex compliance frameworks, and misguided technology implementation. Many feel like helpless bystanders.

When it comes to advocacy, we’re our own worst enemies. Our efforts are fragmented as evidenced by low AMA membership and infighting between specialties. Our opponents (hospitals, insurance companies, and pharma) have better lobbyists and greater funding. Like me, many physicians are apolitical and have little time or motivation to engage in the political process. Those that do often come away frustrated.

For its part, AAHKS has extensively engaged policymakers on bundled payments, APMs, and other value-focused initiatives. Hip and knee replacement surgeons have been on the forefront of VBC models. Despite these efforts, we’re losing ground. It was against that backdrop that I headed to DC with several of my surgeon colleagues for an intense day of advocacy work on Capitol Hill.

On the Hill

I have a soft spot for the DC area having spent my fellowship year living just outside the city. My wife and I went on our first date in Georgetown and spent many weekends enjoying the city and its surroundings. This time of year, the cherry blossoms are quite the sight, too. But this was a business trip — I don’t wear a suit lightly.

Wandering the Halls

Following a morning strategy session, our contingent divided into groups and set off for the Hill. If you’ve never been, the Capitol complex feels endless and maze-like. Left to my own devices, I’d probably still be wandering around the basement corridors. Unfortunately, I didn’t get a chance to check out the gym (jeans allowed, apparently).

My group began on the House side, starting with one of my own representatives, Seth Moulton who’s from Salem, Massachusetts — one town over from where I live. Rep. Moulton’s office was small and cozy with an old school politics feel. As we sat waiting, a screen cycling through Salem-area landmarks flashed the church where my daughters hold their school Christmas pageant. Small world.

One thing I quickly learned — we wouldn’t be meeting with the Representative himself. Instead, we’d be talking to one of his aides. In fact, that was true of all the meetings we had that day. With the exception of Senator Barrasso from Wyoming (a former Orthopedic Surgeon doing us a courtesy), all of our meetings would be with staff, not the members themselves.

Congressional aides are bright, hopeful twenty-somethings from various backgrounds with various career goals. All were good listeners, but it was clear that their degree of healthcare policy knowledge varied greatly. All nodded politely and took notes, but terms like “the RUC” and “APMs” didn’t appear to register for most — with a couple notable exceptions.

Staffers for Senator Marshall of Kansas and Senator Warren of my home state of Massachusetts engaged us on a deeper level. At the same time, it became clear that Congress thinks about healthcare reform far differently than physicians (or at least Orthopedic Surgeons). In all cases, it felt like there was a massive opportunity for each side to educate the other — assuming both parties are willing.

Which brings me to the E&C Health Subcommittee hearing on provider affordability.

Testifying to an Empty Room

Like most people, I’ve seen a few Congressional hearings on TV. This was my first time attending one in person. A few things stood out.

First, the room was smaller than I imagined, perhaps because this was a subcommittee hearing. The Representatives sat in an elevated, three-tiered gallery while the people testifying sat at a long table slightly below them. There were six witnesses in all with the AMA, AAFP, and AHA all represented. I recognized a couple of them: Elizabeth Mitchell from the Purchaser Business Group on Health and Dr. Anthony DiGiorgio, a UCSF Neurosurgeon who posts frequently on X. Much of the discussion closely mirrored our own talking points.

The next thing that struck me was just how few of the Reps bothered to show up. Don’t they take attendance at these things? Those who did attend came and went throughout the session. Few stayed for the whole hearing. (Notably, AOC left about halfway through). At one point, the gallery was maybe half full, but most of the time only about one-third of the seats were occupied. Most attendees seemed bored. Meanwhile, the six witnesses spoke passionately and competently.

Once the prepared testimony portion concluded, it was question time. Reps reappeared to claim their five minutes while staffers scurried to prop up visual aids. What followed was largely four-and-a-half minutes of grandstanding and 30 seconds relevant discussion. Their talking points entered into the record, most of the Representatives again disappeared into the bowels of Congress.

I left the hearing both disappointed and discouraged. An opportunity to engage in meaningful discussion with knowledgeable people was squandered in the name of political theater. The three feet between the witnesses and the members of Congress might as well have been a mile. They weren’t even close enough to talk past one another.

Perhaps I’m being a bit harsh here. To be fair, there was a lot going on at the time (Iran, the TSA/DHS kerfuffle). Maybe this is simply how the process works. Still, it seemed…highly unproductive. I felt for the six witnesses who did prep work and took time out of their schedules only to testify to a half empty room of bored people.

I get it, healthcare policy can be dry. I’m sure stakeholders’ self-serving grievances blend together after a while. Healthcare is so complex that trying to solve it can feel futile. But you have to start somewhere.

Our Talking Points

Narrow, well-defined lobbying efforts tend to be more effective than broad agendas. We came to DC with a list of three main talking points:

• The effect of Medicare reimbursement cuts

• Continued support for alternative payment model (APM) participation

• Support for prior authorization reform and the Improving Seniors Timely Access to Care Act (ISTACA)

My main focus here will be the first two which deserve deeper discussion. The final point, prior authorization reform, is the least controversial and enjoys wide bipartisan support. An official analysis of the ISTACA found that the bill would be cost neutral, a rarity in healthcare.

Medicare reimbursement cuts are a stickier issue. The current administration and MAHA movement favor primary care and prevention, at the expense of specialists (thanks in large part to budget neutrality). Orthopedic Surgeons are no strangers to APMs. In fact, they’re the clubhouse leaders in VBC participation. But newer CMMI models also favor primary over specialty care.

That approach may be misguided.

Medicare Cuts to Hip and Knee Replacements

Over the last 20 years, inflation-adjusted physician reimbursement has fallen 33%. During that same time period, Medicare payments for hip and knee replacements are down 55%.

Physician reimbursement accounts for less than 10% of the cost of lower extremity joint replacement. In the face of an aging, less healthy population, surgeons have improved outcomes and reduced costs. Contrary to popular belief, they’ve taken on more chronic care management and care coordination duties. Surgeons have safely shifted care to lower-cost facilities and drastically reduced utilization of expensive post-acute care.

Despite creating significant value for patients and CMS, hip and knee surgeons have faced numerous cuts, all exacerbated by a lack of annual inflationary updates. Hospitals get them, docs don’t.

Here’s the ledger:

• 5.4% cut to hip & knee replacement (2021)

• 2% Medicare Sequestration (April/July 2022)

• 10.9% conversion factor reduction (2021-2025)

• 5% reduction to the APM incentive payment (2026)

• 10.5% total cut to hip & knee replacement, including a 2.5% “efficiency” cut (2026)

• A proposed 5.5% “revaluing” cut for 2027 (the third such revaluation in recent years)

The inflation-adjusted decline in Medicare reimbursement for THA and TKA is well documented. As the numbers show, joint replacement surgeons are being hit from multiple angles. Between 2000 and 2024, inflation-adjusted Medicare reimbursement decreased 57% for TKA and 54% for THA — the largest decrease of all high-volume inpatient procedures.

If trends continue, the situation could be even more dire soon. A provocative 2024 study suggests that, in a worst-but-plausible-case scenario, joint replacement reimbursement could fall below minimum wage by 2030. Using a three-step mathematical model, the authors projected THA reimbursement to land at $93.78 ($14/hr.), TKA reimbursement at $101.55 ($15/hr.). Step aside AI doomerism.

Chris Klomp, a rising star in the current healthcare administration, wants less complaining about reimbursement and more “deflationary” forces. It doesn’t get more deflationary than a 55% reduction over 20 years. Klomp, who comes from a health tech/private equity background, is looking for big ideas that fit into the broader “Make America Healthy Again” ethos of prevention, improved health span, and tech-enabled, holistic care.

It seems clear that a simple “we deserve more” argument is likely to be a non-starter. That’s fair, and I largely agree. I’ve written extensively in the past about innovative MSK care models like The Hub-and-Spoke Model, the ATV Model, and other outside-the-box ideas. As discussed below, I’m dubious of CMS/CMMI’s approach but appreciate the continued iteration.

I (and many others) have also argued that physician payment deflation paradoxically leads to broader healthcare inflation via consolidation. Shaking physicians down for pocket change seems counterproductive. If the administration's goal is actually deflationary healthcare, consolidation is the enemy, and physician payment compression is one of its primary drivers.

In the face of declining reimbursements, joint replacement surgeons have long threatened to drop Medicare. CMS has repeatedly — and successfully — called this bluff. That too might be changing as opt out numbers slowly creep up. Access will suffer, and the cost curve will continue moving up and to the right.

Finally, I’d argue there are several easy wins being left on the table. Enacting site-neutral payments, repealing CON laws, lifting the moratorium on physician-owned hospitals, and reforming Stark laws and the Anti-Kickback Statute would immediately level the playing field. Ignoring these issues while squeezing physician payments is at best misguided.

Support Value-Based Care Through APMs

Orthopedic Surgeons get a lot of flak when it comes to overutilization and low value care. MSK procedures are big ticket items that cost employers and the government (i.e., taxpayers) a lot of money. The “volume over value” and “unnecessary care” tropes ignore an important truth: hip and knee replacement surgeons have been the biggest participants in alternative payment models.

Unfortunately, they’ve also been canaries in the coal mine; a cautionary tale of what happens when you do everything CMS asks and still see your reimbursement cut in half.

AAHKS and its surgeon members have repeatedly engaged CMS on VBC initiatives like CJR, BPCI, and BPCI-A, working to improve the models while ensuring patient access. The effectiveness of this work is backed by evidence. Studies show that physician-led models ($855/episode cost reduction) perform better than hospital-led ones ($613 reduction). Furthermore, a recent analysis of BPCI-A found that physician group practices accounted for 73% of all joint replacement episodes.

Despite the superior value offered by surgeon-led bundled payment programs and their willingness to participate, CMS seems determined to cut specialists out of the loop. CJR and BPCI have been replaced by TEAM, a model that squarely puts hospitals and health systems in control. While the model allows for gainsharing, offering such agreements is fully at the discretion of the participating facility.

The Ambulatory Specialty Model (ASM) notwithstanding, other recent vintage CMMI models follow a similar playbook. ACCESS creates a tech layer designed to route patients around specialists. ACO LEAD puts PCPs in control, allowing them to serve as conveners of surgical episodes with CMS’ help (and data). Meanwhile, specialists are grouped into narrow networks based on “shadow bundles” with limited visibility or control.

Chris Klomp’s counterpart at CMMI, Abe Sutton, also comes from a venture capital/health tech background. While Klomp wants big ideas, not grievances, Sutton appears to be taking the stance that specialists should be passengers, not drivers. His thesis seems to be that value is created by avoiding specialty care whenever possible, preferably by using tech.

Tech-enabled avoidance only works to a point. Steering patients away from appropriate specialty care comes with unintended clinical and cost consequences. Delayed or avoided joint replacements in appropriate patients become falls, functional decline, opioid use, and eventually higher-acuity surgery. That’s low value care by any definition.

I’d argue that Klomp and Sutton are underestimating the role Orthopedic Surgeons can play. Joint replacement surgeons have embraced the big ideas Klomp favors including site of service migration, direct contracting, and reduction in postacute care. They’ve proven to be more than capable of leading the value-focused initiatives Sutton prefers.

The simple truth may be that CMS/CMMI learned, when it comes to joint replacements, FFS is VBC.

Leaving DC

My DC trip was a firsthand, eye-opening look at the political process. What I learned isn’t surprising: politics is slow, imperfect, and a bit absurd. I suppose the same could be said of healthcare delivery. Maybe physicians and members of Congress have something in common after all. One thing is clear: Orthopedic Surgeons have a lot to learn about advocacy and politicians have a lot to learn about what’s actually happening in healthcare.

I don’t envision myself ever becoming a politician. That hasn’t changed. But I could see myself doing more advocacy work, perhaps even as a lobbyist. I’d probably approach it from a slightly different angle, functioning closer to an expert witness than someone trying to trade access for preferred treatment.

Is there a place for someone who shows up with data, lays bare his own motives, and makes merit-based arguments?

We all have biases. But honest advocacy grounded in data and a genuine interest in solving hard problems could be a net positive. It’s one thing to believe strongly in the position you’re taking. It’s another thing to bring evidence to back your opinions. Refuting data to protect your position doesn’t make it untrue. Maybe Klomp and Sutton, with their private sector backgrounds, are more receptive to a straightforward approach.

As a pragmatist, I’m dubious that my 36 hours in DC changed a whole lot. I’d still go back.

Subscribe now

Ben Schwartz, MD, MBA

Editor-in-Chief

The issue is that the echo chamber has declared “game over” while most people are still looking for their seats. The constant hype, doomerism, and self-promotion are counterproductive, making it impossible to separate fact from fiction. Predictably, backlash has already started. People love an upstart; they sour when plucky becomes arrogant.

No one knows for sure where all this will end up. There's a world in which AI is as revolutionary as some are claiming — maybe more so. There's also a world in which AI is evolutionary, confined to narrow use cases with significant guardrails and human oversight. (Amazon reportedly found this out the hard way).

It's OK to admit we don't know. It's OK to say this may be messy and take a while. There will be (and already are) really cool use cases. There will be failures too, some pretty spectacular (and potentially harmful).

Maybe let's see where things go. Let's get some real-world, proven, validated examples of AI wins, not hyperbolic "what-if" scenarios or fantastical "Tales of the Vibecoders." The more durable a technology becomes, the less you tend to hear about it. Right now, the buzz around AI feels slightly insecure and defensive, especially in healthcare.

Some are no longer pretending that “doctors-who-use-AI” is the ideal state. That was a quaint idea, but I’m not sure everyone really believed it. From some quarters, AI is being pitched as the only thing that can save healthcare. With the knowledge gate blown wide open, access explodes, marginal costs approach zero, and medicine becomes fully democratized. Docs — we’ll call you if we need you.

(Elon Musk recently predicted robot surgeons will surpass the best human surgeons in 5 years. Looks like I timed my career pivot perfectly!)

Still, there are issues that require attention.

Regulation is starting to catch up with the pace of AI innovation. New York state recently introduced a bill prohibiting chatbots from giving professional advice. Other states have taken similar steps, although AI regulation remains messy at the federal level. No one knows how to draw the line between providing information and providing care.

Meanwhile, Utah, in partnership with Doctronic, became the first state to allow AI-enabled prescription renewals. It didn’t take long for Mindgard, and AI security firm, to hack the system with concerning results. Along with tricking the bot into revealing its inner workings, the Mindgard team got it to spread conspiracy theories about vaccines, recommend methamphetamines for social withdrawal, advise users on how to make meth, and triple a patient’s Oxycontin dosage.

Depending on your point of view, this is unfair red-teaming or a warning sign.

The counterargument is that human clinicians are far from perfect — which is indisputably true. Why should we hold AI to impossible standards and give humans a pass? It’s a fair but incomplete point. Flawed system or not, human doctors are held to liability and accountability standards. AI isn’t (yet).

Honestly, I’m not sure what to make of health tech “whataboutism.” I see the point, but it feels more like rationalization than validation. AI is supposed to raise the ceiling, not grope for the floor.

I also can’t help thinking there’d be less pushback if there was a little more hubris and a little less contempt. We’re all frustrated with the current state of healthcare. Everyone agrees that healthcare is complicated. But AI is not a panacea.

Docs need to keep an open mind too. However, our skepticism comes from experience, not technophobia. We don’t want to be gatekeepers of medical knowledge; we want people to recognize the nuances of care delivery. Healthcare is not black and white, solvable by the right combination of 1s and 0s. Assuming so is a fundamental misunderstanding of how medicine works.

The more I use AI, the more I believe in it — and the more apparent its limitations become. In healthcare, we can’t skip or gloss over the part where we figure out if it works, who’s accountable when it doesn’t, and what we lose by taking humans — and humanity — out of the loop. As fed up with the system as we are, weaponizing that angst isn’t the answer.

Technology is always at its best when it fades into the background. The less we hear about AI, the more we’ll know we’re on the right track.

Ben Schwartz, MD, MBA

Editor-in-Chief