Medicare pays for the clearance. It does not pay for the optimization.
Every specialist value-based proposal currently on the table depends on the same thing: moving a patient from inpatient-bound to ASC-eligible before the operation. That work is worth roughly $9,000–$10,000 per case in avoided Medicare spend. No billing code captures it. So the work that produces the savings is the work nobody is paid to do.
Clearance is one visit. Optimization is a program.
These are routinely treated as the same thing, and they are not remotely the same thing. The confusion is why the second one doesn’t happen.
One encounter. The physician documents current status, stratifies risk, and sends a note. The patient arrives as they are, has the case cancelled, or gets a rushed attempt at bare-minimum mitigation.
Billable.E&M coding fits it correctly.
Multiple visits. Measurable targets — BMI, A1c, smoking cessation, blood-pressure control. Active management of modifiable conditions, with documented progress communicated back to the operating surgeon.
Not billable. No code describes it.
“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.”
The savings are real, and they are large.
Roughly 30–40% of Medicare primary total joint arthroplasty is currently appropriate for an ASC, and that share is estimated to expand toward 45–50% through preoperative optimization. Each patient who moves is the difference between an inpatient episode and an ASC episode.
There is also evidence the selection effect is measurable. One 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% ASA class 3 or higher versus 41.7% before. The healthier patients moved. The question is whether the rest can be moved deliberately rather than sorted passively.
And the health gains appear to persist. Patients who stop smoking before surgery are more likely to remain smoke-free afterward; perioperative glycemic improvements show similar durability. A program built to make someone safe for an operation produces chronic-disease outcomes as a side effect — the same outcomes other federal programs are spending heavily to chase directly.
Both proposals pay for it. They disagree about how.
These get discussed as if they were one idea. They resolve the appropriateness problem in genuinely different ways, and the difference decides who bears which incentive.
The physician group holds episode accountability across ASC, HOPD and inpatient, with a co-management payment to the primary care physician for documented optimization visits, plus an outcome bonus paid out of the surgeon’s own savings if the patient clears the 90-day window.
The operation still has to happen for the episode to exist. Appropriateness is handled through risk stratification, stop-loss, and the fact that a complication lands in the surgeon’s own reconciliation.
Paid to manage osteoarthritis for a year whether or not surgery happens, held accountable to patient-reported outcomes. Not capitation — withholding care shows up in the PROMs.
Reported result: a surgery rate roughly 20% lower per capita at equal disease severity. Payment is decoupled from the knife entirely, which is a stronger answer to appropriateness and a harder sell to a fee-for-service practice.
Both require an instrument neither one describes.
A structured optimization program is only real if it produces a record. “Documented progress communicated back to the surgeon” is the load-bearing phrase in the episode proposal, and “accountable to patient-reported outcomes” is the load-bearing phrase in the condition-based one. Both describe an artifact: a dated, longitudinal account of what was tried, for how long, and whether function actually changed.
That artifact has one hard design requirement, and it is easy to get wrong. It has to be symmetric. If treatment worked, that must show as plainly as if it failed. A record that only pays off when someone gets an operation is not an optimization instrument — it is a steering tool wearing one’s clothes, and it will be read that way by anyone evaluating it.
Everything else follows from that. Patient-held rather than vendor-held. Portable across whichever payment model actually arrives. Useful to the patient on its own terms, so it gets maintained whether or not a policy ever pays for it.
Comments on CMS–1848–P close September 14, 2026.
The CY2027 Physician Fee Schedule proposed rule cuts arthroplasty work RVUs materially while raising the ACO longitudinal-care add-on — value moving out of the procedure and toward whoever manages the relationship around it. The specialty’s response so far has been to argue, correctly, that surgeons need a genuine path into value-based models.
The argument that is not being made is the specific one: the mechanism those models depend on already has a name, a duration, measurable targets, and published evidence of durable benefit — and it has no payment pathway. A comment that names the mechanism is more useful to the agency than one that names the grievance.
Sourcing note.BPCI-X is a proposal published by a specialty editor; it carries no CMS status, and its financial figures are explicitly labelled illustrative by its author. The condition-based bundle described here is operating and its result is self-reported. The arthroplasty RVU changes and the ACO modifier are from the proposed rule itself. Treat each accordingly — and verify the underlying studies before relying on any of it.
The rest of the rule.
This page argues one narrow thing. Three neighbours carry the parts it assumes.