No autonomous anesthesia system is cleared, and the specialty bills by time. Anesthesiology's income exposure runs through CRNA scope, commercial rates and a separate, lower conversion factor.

In late 2024, Anthem Blue Cross Blue Shield announced it would stop paying for anesthesia time beyond an insurer-set limit in Connecticut, New York and Missouri from February 1, 2025, regardless of how long the operation ran. The policy was withdrawn on December 5, 2024, before it took effect, after a public reaction that had nothing to do with artificial intelligence. It showed where an anesthesiologist's income is exposed: at the unit of payment, which is time, and at the commercial rate paid for it, which runs three and a half times Medicare or more.
No autonomous anesthesia system has been cleared by the FDA. The specialty's exposure to AI is real and it does not run through the machine at the head of the bed.
Figures are as of October 2026 and the perishable ones will be updated.
The deployed AI in anesthesia is the Hypotension Prediction Index, a monitoring algorithm cleared by the FDA that warns of a likely drop in blood pressure before it happens. It advises; a clinician acts. Closed-loop systems that titrate hypnotic drugs to a processed EEG target were reviewed in a 2026 meta-analysis in the British Journal of Anaesthesia, and none is cleared for sale in the United States. The FDA's list of AI-enabled devices stood at 1,614 in its September 2026 update, which counts authorizations through June 2026, and 1,230 of them were radiology devices. Anesthesia's share is a handful of monitors.
The economic reason this matters more than it would in another specialty is the billing unit. Medicare pays anesthesia as base units for the procedure plus time units for its duration, multiplied by a conversion factor. A tool that makes the cognitive work easier does not shorten the case, and a tool that shortens the case is a surgical tool, credited to the surgeon. The 2026 efficiency adjustment that cut work values on procedural codes by 2.5% exempted time-based services by design. Automation in anesthesia has no billable unit to compress.
Surgical volume grows with the population over 65, and the specialty's demand exposure is to where the cases go. CMS removed 285 procedures, most of them musculoskeletal, from the inpatient-only list for 2026 and finalized the list's complete elimination by January 1, 2028. The same rule added 560 surgical procedures to the ambulatory surgery center list. CMS's July 2026 proposed rule for 2027 would remove 637 more services from the inpatient-only list and add 618 codes to the ASC list. Each case that moves from a hospital to an ASC moves from a facility where anesthesia rates and stipends are negotiated at system scale to one where they are negotiated center by center.
One case class is shrinking outright. In an Epic Cosmos analysis of patients who had a primary sleeve gastrectomy or gastric bypass, published in Surgical Endoscopy in July 2026, bariatric surgery utilization peaked in the fourth quarter of 2022 and fell 39% through the fourth quarter of 2025. The share of those patients dispensed semaglutide or tirzepatide before surgery rose from 0.2% to 35.3% between the fourth quarters of 2018 and 2025. Anesthesia has no say in that substitution; it is upstream of the operating room.
Anesthesiologists do not generate their own volume. The surgeon's schedule and the facility's block time set it. The economic sign of that dependence is the subsidy: the share of surgery-center leaders expecting to pay anesthesia groups a stipend to cover the gap between collections and the cost of coverage rose from 28% in 2024 to 44% in 2025, in VMG Health's surveys, and 67% of the 97 leaders in its latest named anesthesia coverage among their top three financial challenges for 2026. A specialty that must be subsidized to staff a facility is a specialty whose collections do not cover its labor at current rates. The subsidy lasts as long as the facility has no cheaper way to staff the room.
The substitution anesthesiology faces is nurse anesthetists and anesthesiologist assistants, and it is governed by state law, not by the FDA. Twenty-seven states and the District of Columbia have opted out of Medicare's requirement that a CRNA be supervised by a physician, on the ASA's count; Vermont and Ohio opted out in July 2026 and the District in September 2026. Certified anesthesiologist assistants can practice in 24 jurisdictions, with Tennessee and Virginia added in 2025. The BLS counts 54,500 nurse anesthetist jobs in 2025 and projects 59,800 by 2035, a 10% increase, which is slower than the 36% it projects for advanced practice nurses as a group. The supply channel is open and it is widening one legislature at a time.
Anesthesia is paid on its own conversion factor, $20.4976 per unit in 2026 for most anesthesiologists, up 0.88% from 2025. The general physician fee schedule pays $33.4009. The ASA attributes the weak 2026 update to a 2.3% across-the-board cut to the anesthesia conversion factor for practice expense and malpractice costs, which absorbed most of the one-year 2.5% increase Congress enacted in July 2025. That increase expires at the end of 2026, and CMS's July 2026 proposed rule would lower the anesthesia conversion factor 1.38% for 2027, to $20.2143.
Because Medicare's anesthesia rate is low, anesthesia groups depend on commercial payers. A 2020 GAO report collected three estimates of the commercial rate as a multiple of Medicare: 3.67 in a Yale study, 3.46 in an ASA study, and 2 to 7 across six services in a Health Care Cost Institute study, and attributed the gap to market concentration and bargaining power. Every dollar of the specialty's income that rests on that multiple is exposed to whatever compresses it: a payer policy like Anthem's, a No Surprises Act arbitration outcome, or a health system that takes the contract in-house. The Anthem episode failed. The next attempt will be written more carefully.
The specialty is short and getting shorter. Facilities reporting an anesthesia staffing shortage rose from 35% in early 2020 to 78% in late 2022, in a study published in Anesthesiology in 2024. HRSA projects anesthesiology supply at 83% of demand in 2038. Burnout, at 39.2% in the AMA's 2025 data, sits below the physician average, though the AMA notes hospital-based specialties underperformed on three of its five measures. The 2026 Match filled every one of 2,171 PGY-1 and PGY-2 positions.
A shortage supports income while it lasts. It also supports every argument for the substitutes in the section above, and 27 governors have already accepted the argument.
The replacement question has a clear answer, and it is the wrong question. The specialty's income runs through a commercial multiple that a single payer policy can move, a state scope-of-practice map that changes every session, a facility subsidy that measures the gap between collections and cost, and a Medicare unit price that is 61% of the general fee schedule's. None of those is a machine. All of them are visible from public data, and the Specialty Exposure Map scores each of them for anesthesiology and for the other specialties on the map.
For an anesthesiologist paid per unit by an employer or a staffing company, the exposure is on the unit price and the pass-through. How the unit price is set, and why it falls while the work rises, is here. For one who owns a share of the group, the stipend and the commercial contract are the assets, and the question is how long each lasts.
The Income Variable puts anesthesiology beside the other hospital-based specialties in its chapter on who captures the productivity dividend, and its second part covers the ownership forms, surgery-center equity among them, that put a physician on the receiving end of a facility's economics rather than at its cost line. Join the launch list for the book and the free companion tools.
This article is general information and analysis, not individualized medical, financial, investment, tax, or legal advice. See the Disclaimers page for the full statement.
Double board-certified reconstructive surgeon in Austin, Texas. Author of The Income Variable.
About the author
Join the launch list for book updates, new articles, and free companion tools.
By submitting, you agree to our Terms of Use and acknowledge our Privacy Policy.