CMS Has Been Paying Primary Care Doctors Per Patient Per Month for 18 Months. Almost Nobody in DPC Has Noticed.

The federal government has been running a payment model experiment for eighteen months that looks, structurally, a lot like DPC.

The program is called the ACO Primary Care Flex Model — ACO PC Flex for short. It was launched by the Center for Medicare and Medicaid Innovation (CMMI) in January 2025. The basic mechanic: primary care physicians in participating physician-led accountable care organizations receive a Prospective Primary Care Payment, a flat monthly fee per enrolled Medicare beneficiary, instead of billing fee-for-service for each visit. The monthly payment replaces the claims-based cycle entirely. Medicare effectively zeros out primary care claims from the participating physicians, and the PPCP covers them instead.

This week, the AMA hosted a webinar introducing physicians to this model. If you’re in DPC and you haven’t heard of ACO PC Flex, you’re not alone — but the program is worth understanding, because it’s running in parallel with something that happened on Tuesday.

What Happened This Week

On July 15, Congress introduced the Patients First Act of 2026. One of its provisions: a five-year Medicare demonstration pilot that would pay primary care physicians a per-member-per-month supplemental payment. DPC Insider covered it the next day.

That was Tuesday. On Thursday, the AMA held its webinar on ACO PC Flex.

Two separate branches of the federal government, working through completely different processes, landed on the same structural idea — prospective, per-patient monthly fees for primary care — in the same seven days. One is a bill that hasn’t passed. The other is already in its second performance year.

What ACO PC Flex Actually Does

The model works like this. Physician-led accountable care organizations with low prior revenue — meaning practices that weren’t already large, hospital-affiliated revenue generators — can apply to participate. If accepted, the ACO’s primary care physicians receive a monthly payment for each Medicare beneficiary attributed to the practice. That payment covers primary care. Per-visit billing for primary care stops.

The CMS 2026 participation data puts the current scale at 23 participating ACOs covering 359,720 people with Traditional Medicare. In performance year 2025, service areas spanned 37 states, with beneficiaries in all 50 states and Washington, D.C.

The model also includes an Advanced Shared Savings Payment — a one-time upfront payment to ACOs to help them invest in care infrastructure before the model produces savings — and ACOs that outperform their benchmarks share in savings with Medicare at the end of the year.

The target participants are specifically physician-led, not hospital-led. That’s not an accident. The Congressional Budget Office has documented that physician-led ACOs achieve greater savings for Medicare than hospital-led ACOs. CMS is explicitly trying to build on that track record.

Why This Is Different From DPC

The comparison has real limits.

DPC practices have opted out of Medicare billing. Their patients pay memberships directly to the practice. Medicare doesn’t touch the transaction. ACO PC Flex is, emphatically, a Medicare program — a program that requires ACO participation, Medicare enrollment, CMS-approved attribution methodology, and a shared savings arrangement. DPC physicians who’ve opted out can’t participate. Their model exists precisely because they decided not to be in this system.

The payment structure also differs in a meaningful way. DPC memberships are paid directly by patients or employers. The PPCP under ACO PC Flex is paid by CMS. One arrangement removes insurance from the loop entirely. The other replaces per-visit insurance billing with per-patient insurance payment. These are not the same thing, and calling them equivalent would overstate the resemblance.

What they share is the logic: a physician knows at the start of the month how much revenue they’ll receive for each patient. They aren’t optimizing for visit volume. They aren’t coding for RVUs. The payment model doesn’t reward more appointments; it rewards better outcomes within a defined cost envelope.

What the Convergence Signals

When an independent payment model used by a few thousand physicians and a CMMI demonstration program running inside Medicare land on the same basic structure, that’s not coincidence. It’s evidence that the per-patient-per-month approach has moved from alternative to legitimate.

The Patients First Act is still a bill. ACO PC Flex is already running. Both face their own constraints. The Patients First Act hasn’t survived committee. ACO PC Flex is limited to 23 ACOs and physician-led organizations only — it isn’t scaling to every primary care practice in the country.

But two things are true simultaneously. First, DPC practices can’t participate in either program. They’re outside Medicare, by choice, and neither the Patients First Act pilot nor ACO PC Flex is designed with them in mind. Second, the payment model those practices have operated on for a decade is now the model two separate parts of the federal government are actively trying to build into Medicare.

That’s an interesting position to be in. DPC wasn’t waiting for validation. But the validation is arriving anyway.

What This Means

For DPC physicians, ACO PC Flex isn’t actionable today — at least not directly. The application period for PY 2026 has already closed, and the program targets physician-led ACO entities, not individual DPC practices.

But the model is worth watching for a different reason. CMS innovation experiments either expand, shrink, or die based on performance year outcomes. If ACO PC Flex produces the kinds of savings and satisfaction data that early physician-led ACO models have suggested are possible, CMMI will face a decision: iterate narrowly, or look for ways to extend the underlying payment logic to broader populations. What the extension looks like, and whether independent primary care practices could eventually fit into it, is an open question.

For now, the picture this week is unusual. DPC practices developed a payment model outside the system that the system is now testing from the inside. Neither the AMA webinar on Thursday nor the Patients First Act bill on Tuesday mentioned DPC by name. They didn’t need to. The structural argument has already been proven by the practices that proved it first.