Atlantic Health Strategies

Behavioral Health Prior Authorization: What Operators Are Actually Fighting in 2026

Table of Contents

Ready to See Results?

From strategy through execution, Atlantic Health Strategies integrates compliance, operations, and growth into durable, measurable results. Let’s put our expertise to work for your organization.

The denial that started this post

Last month a 24-bed Level 3.5 program in Palm Beach County sent us a stack of 31 concurrent review denials from a single national payer, all issued inside a 60-day window. Average length of stay approved on initial cert: 4.2 days. Medical necessity criteria cited: the payer’s internal guidelines, not the ASAM Criteria, 4th Edition. The clinical director had been writing peer-to-peer notes at 9pm three nights a week for a quarter.

That is the actual operating reality of behavioral health prior authorization in 2026. Not a policy abstraction. A clinical director burning out while $480,000 in billed charges sits in suspended AR, and days in AR creeping past 71. The payer knows the math. Every day a claim sits unworked is a day closer to the operator writing it off.

Prior auth is not a paperwork problem. It is a margin transfer mechanism, and the federal government has finally, partially, started to say so out loud.

What the DOL and CMS actually did (and did not do)

Behavioral Health Prior Authorization: What Operators Are Actually Fighting in 2026 — What the DOL and CMS actually did (and did not do)

The MHPAEA final rule from the Department of Labor, HHS, and Treasury, published September 2024, requires plans to perform comparative analyses on every nonquantitative treatment limitation, including prior authorization, concurrent review, and network composition. The DOL’s FY2024 MHPAEA enforcement report flagged behavioral health prior authorization as the single most common NQTL violation across the audits it closed. That is not a footnote. That is the regulator telling you where to aim.

CMS followed in January 2024 with the Interoperability and Prior Authorization final rule (CMS-0057-F), which forces Medicare Advantage, Medicaid managed care, and CHIP plans to issue standard prior auth decisions within 7 calendar days and urgent decisions within 72 hours, starting January 1, 2026. Public reporting of prior auth metrics begins in 2026 as well. Commercial ERISA plans are not directly covered, which is the gap every operator should understand.

What the rules did not do: they did not ban concurrent review for residential SUD treatment, they did not mandate ASAM 4 as the sole medical necessity standard, and they did not give you a private right of action. Enforcement still runs through DOL audits, state AG actions, and ERISA litigation.

Where state regulators are picking up the slack

New York’s Office of the Attorney General settled with UnitedHealthcare in 2024 for $13.6 million plus restitution over behavioral health denials, building on the 2021 New York and federal $15.6 million settlement. OASAS and DFS in New York have continued to push payers to use ASAM Criteria for SUD level of care decisions, and as of the 2023 statutory update, commercial plans in New York must defer to ASAM for SUD medical necessity. California’s DMHC issued enforcement actions in 2023 and 2024 against multiple plans for parity violations, with penalties in the seven figures. Illinois passed HB 2595 years ago requiring the use of generally accepted standards; enforcement has been uneven but not absent.

Florida is a different animal. The Office of Insurance Regulation has not been aggressive on parity. AHCA’s focus on the Medicaid managed care side has been fraud, waste, and abuse, not prior auth abuse by the plans. If you operate in Florida and your appeals strategy assumes a regulator will eventually back you up, rebuild the strategy. Your pressure points are contractual and clinical, not regulatory.

What to actually change on Monday

Five operational moves that move the denial rate. First, every initial cert and concurrent review note must cite the ASAM Criteria, 4th Edition by dimension and by level of care, in the clinician’s own words, in the medical record before the call. Not after. Payer reviewers are trained to ask for specifics; vague risk language gets you a denial and a peer-to-peer you will lose.

Second, track NQTL evidence per payer. Pull a sample of 50 concurrent reviews per quarter per payer and log: criteria cited, reviewer credentials, decision turnaround, overturn rate on appeal. That dataset is what a DOL investigator or a plaintiff’s attorney will ask for. It is also what gets you a meaningful conversation at contract renewal. Third, for ERISA self-funded plans, send a written request for the comparative analysis under the 2024 final rule. Most plans will not have one ready. That silence is useful.

Fourth, file external appeals. The IRO overturn rate on behavioral health denials runs above 40 percent in several states based on DFS and DMHC data. Operators who appeal less than 25 percent of denials are leaving real money on the table. Fifth, segment your payer mix on the rate sheet by denial behavior, not just by allowable. A payer paying 92 percent of Medicare with a 22 percent denial rate is worse than a payer paying 78 percent with a 6 percent denial rate. Run the math. Renegotiate or terminate.

Behavioral Health Prior Authorization: What Operators Are Actually Fighting in 2026 — What to actually change on Monday

Where this is heading, and a note from Amelia Island

The 2026 CMS public reporting requirement is going to embarrass several large plans. Expect the data to feed a new wave of state legislation in 2026 and 2027, particularly in states with active AGs. Expect commercial plans to quietly tighten initial cert criteria before the parity comparative analyses get subpoenaed. Expect more single case agreements offered to out-of-network programs that have documented their medical necessity better than the in-network competition.

AHS is sponsoring the Women in Leadership Luncheon at NAATP National in Amelia Island this week, May 4 through 6. Allison, Benjamin, Leah, and Sariah will be there. If you want to walk through your denial dataset, your payer mix, or how to actually structure the comparative analysis request letter, find one of them. Bring a denial log if you have one. The conversation is more useful with real numbers in front of us.

Request a Free Consultation

Scroll to Top