Most behavioral-health clinicians have never heard of CMS-0057-F. They are about to.
On January 1, 2026, the rule's decision-timing requirements took effect: urgent prior-authorization requests must be decided within 72 hours, standard requests within 7 days. Late this March, the harder edge lands — every Medicare Advantage organization, Medicaid managed-care plan, CHIP managed-care plan, and federally-facilitated marketplace QHP issuer subject to the rule has to publicly report prior-authorization metrics on their website. That includes denial rates, broken down by service category, refreshed annually.
For behavioral health, this is not a small change. It changes the math.
Why this matters more for BH than for other specialties
Behavioral-health claim denial rates run 16–22% in commercial and Medicare Advantage books. The KFF's 2024 analysis of MA prior authorizations found BH services were denied at roughly double the rate of medical/surgical services, and Johns Hopkins' October 2025 review of payer data described BH parity enforcement as “structurally weak.” Until now, the only people who knew this in any payer-specific detail were the payers themselves. Practices saw their own pain; nobody could see the system.
Once denial rates are published per payer, three things change at once:
- Practice leverage in contracting. A payer's BH denial rate becomes part of every contract negotiation. A network with a 3% denial rate is no longer indistinguishable from one with 22%.
- Patient steering. Patients (and the employers who hire them) will start asking questions a payer can no longer dodge with “our utilization-management process protects members.”
- Regulator attention. Parity enforcement under MHPAEA has been chronically under- resourced. Public data makes selective enforcement much harder.
The leverage shift is not automatic
Just because the data becomes public doesn't mean a small BH practice automatically benefits. The payers will lawyer their categorization (every claim moved into a different “service category” is a denial that doesn't count in the headline rate). The published data is yearly, not real-time. And the rule applies to prior-authorization denials specifically — denials after claim adjudication (CARC codes) sit in a separate, still-opaque universe.
What this means practically: the public data will be most useful as a directional signal— “this payer's BH denial rate is twice the network average” — and most actionable when paired with the practice's own claim-level data. Practices that have been tracking nothing will not suddenly be empowered by a published number. Practices that already know which payer denied what, why, and how often will be able to act on the disclosure within hours.
What a behavioral-health practice should do in the next six months
Three things, in order:
- Start tracking your own denials with structure. Most practices know they have denials. Few can pull a report grouped by payer × CARC code × CPT code. Without that, you have nothing to compare to the public data. Bonus: tracking creates appeals you wouldn't otherwise have written.
- Tighten documentation to known payer criteria. Most BH denials are not about whether the patient needed care — they are about whether the note shows it. Each major payer publishes its medical-necessity rules; map them to your note template so the documentation a denial would ask for is already in the note before sign.
- Build the appeal habit, even small ones. Roughly half of appealed BH denials get overturned. Practices that never appeal teach payers that denying the claim is free. Practices that appeal — even on a fraction of denials — teach payers the opposite. The asymmetry compounds over a year.
Where AI shifts the floor
The reason practices don't do all three is staff time. Appealing one claim takes ~$55–$150 of someone's hour, and the practice owner usually decides — correctly, in isolation — that the math doesn't work on a single $130 90834. The economics flip when the documentation arrives already-appeal-ready and the appeal letter draft is sitting on the screen the moment the denial lands. That is the part we've been building at Trellova — an ambient AI scribe that produces a source-linked clinical note, a denial-recovery layer that drafts appeals citing the payer's own criteria, and a closed loop the practice owner can actually run without hiring a billing manager.
The disclosure rule doesn't fix BH denials by itself. It changes the conversation. Practices that spend the next six months getting their own data house in order will be the ones in a position to use the disclosure when it lands.
Trellova is building the operations layer for 2–10 clinician behavioral-health practices. If you're thinking about how to get ready for March 2026, we'd love to talk — get in touch.