Physicians in roughly a quarter of the country's metro areas will soon be scored individually on quality, cost, improvement activities and interoperability

In February, the Centers for Medicare & Medicaid Services (CMS) published a list of 6,637 physicians selected for the Ambulatory Specialty Model (ASM), including their names, National Provider Identifiers (NPIs) and organizations. The preliminary list was built using 2025 claims data, with the final participant list expected soon.

In June, a Dayton, Ohio, neurosurgeon speaking at a spine conference described searching the list for colleagues. He wasn't on it, but plenty of people he knew were, and they had no idea.

That should get the attention of every health system employing specialists in the markets affected by ASM, because their Medicare reimbursement is already on the clock. Beginning Jan. 1, 2027, specialists treating heart failure or low back pain in approximately a quarter of the country's metropolitan areas will be scored individually on quality, cost, improvement activities and interoperability. Cost alone will only account for half of the score.

This isn't just another reporting requirement. Physicians will be measured against other participants in their cohort, meaning the benchmark doesn't exist until everyone's performance is known. Physicians perform and report in 2027; CMS calculates the benchmark in 2028. As it is written today, this calculation is done in a way that someone will be at the bottom. And payment adjustments arrive in 2029. Those adjustments can range from -9% to +9%, eventually widening to plus or minus 12%. The adjustment applies to all of an individual physician's Medicare Part B payment. Not just heart failure or low back pain visits. Everything they bill.

The risk has moved from the organization to the physician

I spent 15 years managing physicians, and the one constant was volatility. The Medicare conversion factor changes; reimbursement changes. As a result, organizations are always budgeting against numbers they don't entirely control.

Historically, much of that risk lived at the organizational level, where health system leadership could manage around it. Under the Merit-based Incentive Payment System (MIPS), organizations could report as a group, allowing physicians under the same taxpayer identification number (TIN) to share a score and payment adjustment. A lower-performing physician's results could be absorbed by stronger performance elsewhere in the organization.

ASM changes the equation by pushing considerably more performance risk down to the individual physician. Participants are scored at the individual TIN/NPI level, with no group reporting option. A cardiologist whose utilization looks perfectly reasonable in an organizational dashboard could be managing a patient population generating substantially higher episodic costs than peers. Those costs may be driven by referral patterns, site of care, access, care coordination or any number of factors the physician doesn't control.

Take the cardiologist treating heart failure. By the time that patient is sitting in her exam room, a primary care physician has already decided when to refer, and how late. An emergency department may have already admitted the patient once, or twice, before anyone called a cardiologist. Imaging was ordered by someone else. If the patient ends up in a skilled nursing facility or back in the hospital 30 days later, that cost lands on the episode too, generated by decisions made in a discharge planning meeting she wasn't in.

She is the specialist. She is not the only author of this cost. ASM scores the whole episode, but it doesn't ask who was responsible for which part of it.

The group average used to obscure those differences, but ASM won't. Because performance is relative, managing everyone toward an organizational average won't be enough. Someone will still land at the bottom of the distribution.

Individual accountability isn't the same as individual control

Not every physician is starting from the same place, because neither are their patients. For example, two cardiologists treating heart failure can produce very different results. One cares for a population with reliable transportation, easy access to primary care, stable housing and the ability to follow a low-sodium diet. The other's patients may rely on a corner store for groceries, struggle to get to follow-up appointments or face financial barriers that make medication adherence difficult. Both physicians can practice excellent medicine, but their cost and quality outcomes may still look very different. That matters when payment models are designed around performance.

The goal of paying for better outcomes is the right one, but health care outcomes are shaped by far more than what happens between a physician and a patient in an exam room. Socioeconomic circumstances, access to food, transportation, housing stability, health literacy, caregiver support, and access to primary and specialty care can all influence whether a treatment plan succeeds. When physicians are compared with peers, health systems need to understand how much of the variation reflects clinical performance, and how much reflects the populations physicians have been asked to serve. Otherwise, we risk creating a perverse incentive: physicians caring for the patients with the greatest needs may also face the greatest performance headwinds.

Physicians can't manage their way out of this alone

We already ask a lot of our physicians. We build their schedules around templates, give them work relative value unit targets, ask them to conduct peer-to-peer calls with payers to justify care for patients sitting in front of them. Now we're adding episode costs, quality measures and performance rankings against physicians they've never met, with as much as 12% of Medicare reimbursement eventually riding on the result. And many of the factors determining that result — panel composition and acuity, access to appointments, imaging utilization, coordination with primary care, referral patterns — sit outside the physician's direct control.

That makes ASM as much an operational challenge for health systems as a performance challenge for physicians. A physician will ultimately be graded based on how the system around them operated years earlier. By the time CMS delivers the score, many of the opportunities to change the outcome will be gone. A physician will be graded in 2029 on how the operation ran in 2027. That's why these questions are worth asking now, while the operation, and the outcome, are still ours to shape: Do the physicians selected for ASM even know they're participating? Can we describe an individual physician's panel, access, episode costs, referral patterns and quality performance without averaging those results across a group? Can we identify which physicians are caring for disproportionately complex or socially vulnerable populations, and understand how those factors may be affecting performance? Can a heart failure cardiologist see their episode costs today, or will they first learn about them when CMS reports the results after the performance period has ended? And if we can't answer these questions yet, are we prepared to tell a physician their Medicare penalties have come before we can tell them why or what to do about it?

Building visibility

CMS is preparing to evaluate physicians individually. Most of us still manage physician performance collectively. We are moving from the team sport of health care to individual contributors. ASM is forcing that mismatch into the open, and closing it means understanding what is happening at the individual physician level while keeping the context to explain why it is happening, which means connecting clinical outcomes with utilization, cost, referral behavior, patient access, panel complexity and the social factors influencing health. The objective isn't to find someone to blame. It's to find where we can still intervene, while there's time left to change the outcome.

ASM may begin with heart failure and low back pain, but it would be foolish to view this as an isolated Medicare experiment. If individual accountability for specialty care produces the savings CMS expects, the framework could provide a model for managing other high-cost chronic conditions.

Today, health systems have a choice. Wait for CMS to deliver the score in 2029, or know the answer before the performance year that decides it even begins. That means taking a closer look at how physician panels are balanced, not just by patient volume, but by capacity, demand and case complexity. It requires understanding how physicians manage their workloads, where imbalances may be affecting performance, and whether current workflows support the outcomes physicians need to achieve. With that visibility, system leaders can have more meaningful, data-informed conversations with physicians and adjust early, while there is still time to influence results.

Holly Burke-Wirth, MHPA, FACHE, is senior vice president of practice management and physician solutions at Health Catalyst. A health care leader with deep experience in provider management, physician group operations and clinical innovation, Burke-Wirth has spent her career helping health care organizations align their operations with the needs of physicians and patients. At Pulse Heart Institute, she led the largest cardiovascular provider platform in the Pacific Northwest, and at MultiCare Health System, she oversaw clinical innovation and quality, including managing its heart failure program.

For the past decade, Burke-Wirth has brought a practical, provider-centered perspective to rethinking traditional health care operations, especially as the delivery of care continues to shift to the ambulatory setting, and she is skilled in identifying opportunities for sustainable health system growth. She received a master's degree in health policy administration from Washington State University and a bachelor's degree in health care administration from Idaho State University.