Your 2026 raise expires on January 1

The CY 2027 conversion factor goes down

• Qualifying APM participants · CY 2026: $33.5675 · CY 2027 proposed: $33.1693 · Change: −1.19%
• Nonqualifying APM participants · CY 2026: $33.4009 · CY 2027 proposed: $32.8409 · Change: −1.68%

Read those percentages carefully, because most coverage is rounding them wrong. The 2.5% increase Public Law 119-21 granted for 2026 does expire. But statutory updates of 0.75% for qualifying APM participants and 0.25% for everyone else, plus a 0.53% budget neutrality adjustment, claw part of it back. The net landing point is 1.19% and 1.68%, not 2.5%.

If your 2027 budget assumed a 2.5% hit, you have more room than you planned for. If it assumed flat, you do not.

What it does to your claims: every claim you submit in 2027 is worth slightly less than the identical claim in 2026. That sounds like a finance problem. It is also a coding operations problem, for a reason most coverage skips.

A payment cut raises the cost of every denial

Work the arithmetic from your own side of the desk.

When reimbursement per claim falls and your cost to rework a denied claim stays flat, the rework becomes a larger share of what the claim is worth. Your denial rate can hold steady and your margin on denied-then-reworked claims still compresses.

Now stack what Kodiak Solutions reported for the first half of 2026. Takebacks rose to 1.57% of accounts receivable from 1.38%. Appeal recovery fell to 2.21% from 2.48%. So you are recovering less per appeal, on claims worth less, at the same cost to chase.

Three places that math shows up in 2027:

Rework cost per claim. If you have never calculated it, do it this month. The number you need is fully loaded coder time per reworked claim, and most teams find it higher than they expected.

Appeal thresholds. Some practices set a dollar floor below which they write a denial off. A lower conversion factor moves claims below that floor. Check whether yours still sits in the right place.

The value of prevention. Every denial you prevent is worth more in 2027 than the same prevention was worth in 2026, because recovering it afterward got more expensive and less likely.

• Physician practices - This hits your top line. Model it before January on your actual Medicare mix rather than the headline percentage.
• RCM and billing companies - Percentage-of-collections contracts absorb the cut directly. If your coding cost per chart is fixed and your revenue per chart is tied to collections, your margin moves without a single operational change.
• Hospitals - Professional fee billing takes this. Facility payment runs on a separate rule.
• ASCs - Your facility rates come from the ASC payment system, but your anesthesia and professional components sit here.

When margin compresses, look at where your certified hours go rather than what they cost. Ask what share landed on charts that never needed certified judgment.

Clean, well-documented, single-condition encounters occupy the same queue position as the complex chart that genuinely needs a credentialed coder. That is the routing problem, and it is measurable in your own data this week.

CODIO AI sends complete charts through AutoPilot for fully autonomous coding and holds thin or contradictory documentation in CoPilot for a certified coder. Teams running it see 45% more output per coder, measured against a manual baseline: same team, same charts, before and after. MediCodio AI also reports cutting coding costs by up to 60–70%.

The lane starts closed. It widens as accuracy is proven on your charts by your auditors, and you can narrow it again at any point.

Calculate your fully loaded cost to rework one denied claim. Divide it by your average Medicare allowed amount for your top five CPT codes. That ratio is your real exposure to the 2027 cut, and it is the number to take into your January budget conversation.