For three decades, the work relative value unit (wRVU) has been the common language of physician compensation. It translated a physician's effort into a single number that organizations could pay against and defend. That common meaning is now being pulled in two directions at once, by a federal payment change that is already in effect and by the early arrival of artificial intelligence (AI) into clinical work. Neither force alone breaks a wRVU-based plan. Together, they raise a question most physician compensation plans are not yet built to answer.
The Efficiency Adjustment
The first change took effect on January 1, 2026. In the 2026 Medicare Physician Fee Schedule (MPFS), the Centers for Medicare & Medicaid Services (CMS) finalized an "efficiency adjustment" that reduces the wRVU value of nearly all non-time-based services by 2.5%, on the rationale that these services are delivered more efficiently over time. Evaluation and management visits are exempt, so the effect lands on procedural, surgical, and diagnostic specialties.
The compensation consequence is direct. Because many physician contracts pay on wRVUs, a lower wRVU value means lower reported productivity for the same clinical work. A coalition of medical organizations made the point to Congress: the reductions decrease physician compensation "despite no reduction in actual work performed."
On its own, this is a manageable problem. The cut is known, it applies on a known date, and a compensation committee can model it. It is the second force that is harder to see.
What AI Changes
While the fee schedule lowered the value of a unit, AI documentation tools began changing how many units a physician can produce.
Ambient AI scribes, which draft clinical notes from a recorded visit, have moved quickly from pilot to routine use, and early evidence ties them to higher measured productivity. A 2026 study in JAMA Network Open, covering nearly 1.2 million encounters at one academic system, found that physicians using an AI scribe generated 5.8% more in RVUs than non-adopters. A separate Mass General Brigham study reported a 12% gain in wRVU-measured productivity.
These two forces are distinct, and they do not simply add together. The efficiency adjustment lowers what a unit is worth. AI, where it has an effect, may change how many units are produced. They can move a physician's reported numbers in opposite directions. What they share is timing: both are altering wRVU figures now.
The Benchmark Lag
The more serious issue sits underneath both changes, in the data used to test compensation for defensibility.
Fair market value (FMV) opinions and the compensation plans built on them rely on national benchmark surveys to establish the market rate per wRVU. These surveys look backward by design. The compensation data available in any given year reflects production from one to two years earlier. In practice, the benchmark behind a compensation plan describes the market as it stood one to two years before the plan was set. For most of the wRVU's history that lag was immaterial, because the market moved slowly enough that the gap did not change the conclusion.
That assumption no longer holds. A compensation plan set in 2026 may measure a physician's current pay against benchmark data that predates both the efficiency adjustment and any AI effect on productivity. Every figure can be accurate and correctly sourced, and the comparison can still be testing two different periods against each other. The risk is not a wrong number. It is a frame of reference that has quietly gone stale.
A Precedent Worth Remembering
This kind of misalignment has occurred before, and the field has a documented response.
In 2021, CMS increased office-visit wRVUs, and the available survey data had not yet caught up. Sophisticated organizations did not rely on a benchmark they knew was misaligned. They adjusted the survey data to estimate the effect of the change. The 2021 change moved in the opposite direction of the 2026 cut, but the mechanism is the same. When a payment change moves the wRVU and the surveys have not caught up, an organization can account for the gap deliberately rather than wait for the data to confirm what it already knows.
That reframes the benchmark lag from an unfixable data limitation into an analytical task, one with a track record, and one that separates a plan that can explain itself from a plan that assumes no one will ask.
Preparing for the Question
The relevant question is not whether your wRVU rates were sound when they were set. It is whether the benchmark behind them still describes the environment your physicians work in today.
Some organizations are already responding, holding internal wRVU values steady against the cut and revisiting their compensation-per-wRVU rates within FMV and commercial reasonableness limits. These are reasonable first steps. They are not a substitute for a compensation plan that accounts for the age of the data behind it. When regulators scrutinize physician compensation, the question is the one that matters most: why do you pay what you pay, and can you show your work? When your compensation plan was last drafted or amended, was the benchmark it relied on already out of date?
Cabra Consulting builds FMV and commercial reasonableness support for exactly this situation, where the unit organizations have long trusted no longer means what it did. If your compensation plan was benchmarked before 2026, it may be worth confirming what it is measuring against.
Notes
[1] American Society of Hematology, "CY 2026 Medicare Physician Fee Schedule Final Rule Summary," Nov 24, 2025 (efficiency adjustment of −2.5% to wRVUs and intraservice time for nearly all non-time-based services; E/M and certain services exempt). https://www.hematology.org/advocacy/federal-rule-summaries/cy-2026-medicare-physician-fee-schedule-final-rule-summary. See also CMS, "CY 2026 Medicare Physician Fee Schedule Final Rule (CMS-1832-F)," Fact Sheet, Oct 31, 2025. https://www.cms.gov/newsroom/fact-sheets/calendar-year-cy-2026-medicare-physician-fee-schedule-final-rule-cms-1832-f
[2] Efficiency Adjustment Delay Act Support Letter (signatory organizations to Reps. Estes and Suozzi), Feb 12, 2026 ("reductions in these values will decrease physician compensation despite no reduction in actual work performed"; reduction effective Jan 1, 2026). https://estes.house.gov/UploadedFiles/Efficiency_Adjustment_Delay_Act_Support_Letter_Final_1.pdf
[3] Holmgren, A.J., et al., "Ambient Artificial Intelligence Scribes and Physician Financial Productivity," JAMA Network Open, Jan 9, 2026 (1.81 more RVUs/week, a 5.8% increase; ~1.2M ambulatory encounters across 1,565 physicians at UCSF Health). https://pmc.ncbi.nlm.nih.gov/articles/PMC12789954/
[4] Mass General Brigham, "Hybrid Ambient Documentation Decreases After-Hours Work, Note Delays for Physicians," Nov 25, 2025 (12% increase in wRVU-measured productivity; published in the Journal of General Internal Medicine). https://www.massgeneralbrigham.org/en/about/newsroom/press-releases/hybrid-ambient-documentation-reduces-after-hours-work