What the AELP 'Ban Prior Authorization' report means for billing teams right now

The AELP report argues that prior authorization should be banned because it costs more than it saves and harms patients. daisyBill's October 1 piece makes the same case for California workers' comp utilization review. Neither will change what lands in your work queue this quarter. What both do is hand billing teams a set of facts, and a set of payer behaviors, that are worth acting on now: appeal more, document the reviewer, treat post-approval denials as a separate case type, and keep the data that makes payers accountable.

Key numbers

• About 650 million hours a year spent on prior authorization nationally, the equivalent of 99,290 full-time physicians and advanced practice clinicians plus 213,474 clinic staff, costing up to $32.7 billion annually (AELP). The 2026 physician shortage is estimated at 84,930 to 96,430.
• 95% of physicians say PA delays care; 79% say it leads patients to abandon treatment; more than one in four report a serious adverse event; 92% report a negative effect on clinical outcomes (AMA 2025 survey, via AELP).
• Nearly nine in ten physicians say PA leads to more utilization, not less.
• Medicare Advantage plans adjudicated nearly 53 million PA requests in 2024, about 1.7 per enrollee, against roughly 628,000 in traditional Medicare, about 0.02 per enrollee. MA cost taxpayers 22% more per patient, about $83 billion, that year (KFF and MedPAC, via AELP).
• 71% of voters in a June 2026 YouGov poll would support legislation barring private insurers from using PA at all.
• UnitedHealthcare dropped PA for 231 procedures in December 2025, out of more than 11,000 billable services, and later pledged a 30% reduction by the end of 2026. Under Texas's gold card law, only about 3% of clinicians qualify.
• California UR historically approves only about 70% of physician-requested treatment. Sedgwick once touted a 54% denial rate as return on investment. IMR overturned 10.2% of denials in 2025 (daisyBill).

What does the report actually argue?

The report's central claim is that insurers use prior authorization to make money, by denying care that is medically necessary and by making each denial expensive enough to fight that most people do not.

The evidence for the second half is the part billing teams will recognize. The report cites Dr. Archelle Georgiou's analysis of 2025 data: 58% of appealed UnitedHealthcare PA denials and 65% of Humana's were overturned. If those rates hold for denials nobody appealed, at $100 a claim, the two plans alone avoid about $100 million a year. Private insurers initially denied 70% of new brand-name prescriptions in 2025, up from 57% in 2021. After appeals, the rate fell to 24%. Most people take the first answer.

The Medicare Advantage comparison carries the cost argument. Traditional Medicare, with almost no prior authorization, processed about 628,000 requests in 2024. Medicare Advantage processed nearly 53 million. MA still cost taxpayers 22% more per patient. If PA saved money, that comparison would look different.

The report also tracks what the industry has offered instead. A nonbinding pledge through AHIP in June 2025. UnitedHealthcare's 231 dropped procedures out of more than 11,000. A Texas gold card program that exempts about 3% of clinicians. AELP's position is that these are too small to matter, and that the 71% of voters who would ban PA outright suggests the political window is wider than the industry pledges assume.

Short of a ban, the report lists conditions any PA program would have to meet: a documented, evidence-based reason for requiring it; an independent adjudicator whose pay is not tied to denials; no AI-based denials; peer-to-peer review by a board-certified peer in the same specialty who discloses credentials; a 24-hour shot clock with automatic approval if the plan does not respond; a standardized electronic process; and a coverage guarantee, meaning an approval must guarantee payment with no retroactive denial.

What does the daisyBill piece add?

Workers' comp. California's utilization review system approves only about 70% of what treating physicians request. One claims administrator, Sedgwick, once marketed a 54% denial rate as a return on investment. Injured workers who dispute a denial go to Independent Medical Review, which overturned 10.2% of denials in 2025. Requests are still faxed. Claims take about seven years to close.

Billing teams that work both commercial and workers' comp will recognize the pattern. The forum differs; the economics do not.

What can a billing team act on today?

Four things, none of which require a new law.

Appeal more

The published overturn rates are 58% and 65% at two of the largest plans. The AMA found nearly 60% of physicians assume they will lose. The gap between those two numbers is money the practice is choosing not to collect. Count denials received against denials worked, by payer, and close the gap. The math for a five-provider practice is in what unappealed denials cost.

Record the peer-to-peer reviewer's name and specialty

AELP's reform list asks for same-specialty, board-certified reviewers who disclose their credentials. Only 16% of physicians say that is what they get today. You do not have to wait for the rule. On every peer-to-peer call, ask for the reviewer's name, specialty and board certification, and write it on the case. When an orthopedic case is denied by a reviewer in an unrelated field, say so in the appeal, and say it in any complaint to the plan or the state regulator. A mismatch you did not record is one you cannot raise.

Treat post-approval denials and recoupments as their own case type

The coverage guarantee is on AELP's list because approvals are not currently guarantees. A service gets authorized, gets performed, and then gets denied or clawed back. These cases are different from a front-end denial. The clinical argument is already won; the payer said yes. The fight is about whether the payer can take the yes back. Build them as a separate case type, with the authorization number and the approval letter as exhibit one, and track them separately. A payer that does this often is telling you something about its approvals.

Contribute to payer accountability data

The industry's June 2025 pledge was nonbinding. UnitedHealthcare's 30% reduction target is self-reported. The only way anyone outside the plans will know whether the pledges held is if practices keep their own numbers: denials by payer, overturn rate by payer, days to decision by payer, and missed decision deadlines. Those figures feed state medical society surveys, AMA data, regulator complaints and, eventually, the kind of report AELP just wrote. Your denial log is the raw material for all of it.

AI on which side of the table?

The report's "no AI-based denials" condition deserves a direct answer from any vendor that uses AI in this space, including us.

UtilReview uses AI on the provider's side of the table, and only there. The model reads the denial, extracts the facts, pulls the payer policy, drafts the appeal and checks the draft against the chart. It does not decide whether care is medically necessary. It does not decide whether a claim gets paid. It does not send anything. A biller or clinician reviews every draft, edits it, and approves it before it leaves the building.

AELP draws the same line. Automation that speeds up a denial is the problem the report describes. Automation that speeds up a well-documented appeal is how a five-person billing team keeps up with a plan that processes 53 million requests a year.

What happens next?

CMS-0057-F's decision timeframes took effect in January 2026 and its API requirements arrive in January 2027. The Improving Seniors' Timely Access to Care Act is still proposed. A ban is a long way off, if it comes at all.

In the meantime, the report's facts are usable today. Most denials can be reversed and most are never challenged, the reviewer is often in the wrong specialty, approvals get taken back, and the payers that behave worst are the ones with the least published data about them. Each of those is a line item a billing team can work this month, without waiting for Congress.