June 5, 2026 · ~12 minute read

Mental Health Parity Appeals After the 2024 MHPAEA Final Rule

I am writing this from a desk in San Diego at 8:40 in the morning, with the dog asleep on my feet and a cold-brew that has gone warm because I keep forgetting it. I read denial letters for a living now. The ones that hit me hardest are still the mental health and SUD ones. A residential admission denied after four days. An IOP denied because the plan wants “outpatient therapy first.” A buprenorphine prior auth that has been pending so long the patient has already relapsed.

My background is in Health Informatics, not law, so my instinct is to look at the data and the rules side by side. With mental health denials, the rules and the data almost never agree. Plans will tell you the same medical necessity criteria apply to mental health as to medical care. Then you look at how they actually apply them. They are not the same. They have never been the same. The 2024 MHPAEA Final Rule was supposed to force the comparison out into the open. The federal government has since pulled back on parts of it. The underlying law is still there.

This post is for people whose plan denied or capped therapy, residential treatment, IOP, PHP, or medication for addiction treatment. It covers commercial plans, ACA marketplace plans, and (with caveats) Medicare Advantage. I will be honest about where parity arguments are strongest and where they are mostly a lever, not a guarantee.

What changed in 2024 and what is actually still in force

The Mental Health Parity and Addiction Equity Act of 2008 (MHPAEA) is the federal law that says group health plans and individual market plans cannot impose more restrictive limits on mental health and substance use disorder (MH/SUD) benefits than on comparable medical and surgical (M/S) benefits. The 2024 Final Rule was published at 89 FR 77586 on September 23, 2024, and it took effect November 22, 2024, with the operational requirements staggered to plan years beginning on or after January 1, 2025 and January 1, 2026.

The big idea in the 2024 rule was non-quantitative treatment limitations (NQTLs). Things like prior authorization, step therapy, network admission criteria, medical necessity definitions, and provider reimbursement methodology. Plans now had to do a written comparative analysis that showed, in writing and in operation, that any NQTL applied to MH/SUD benefits was no more restrictive than what they applied to comparable medical benefits. The rule was codified in new regulations at 26 CFR 54.9812-2, 29 CFR 2590.712-1, and 45 CFR 146.137.

Then in May 2025, the Departments of Labor, HHS, and Treasury announced they would not enforce the new pieces of the 2024 Final Rule while related litigation plays out, plus 18 months after final resolution. That sounds like the whole thing died. It did not.

Here is what is still required, as of mid-2026, and what is paused.

Still required by statute (Consolidated Appropriations Act of 2021):

• Plans must prepare and maintain a written NQTL comparative analysis.
• Plans must provide the comparative analysis to regulators and to participants on request.
• The 2013 MHPAEA regulations at 45 CFR 146.136 are still in force.

Paused under the non-enforcement policy:

• The new 2024 “meaningful benefits” standard.
• The new data collection and outcomes-based requirements that were keyed to plan years 2026 and later.
• The fiduciary certification requirement that was added in the 2024 rule.

For the appeals work I do, the part that matters most was never paused. Plans still have to be able to defend, in writing, why their NQTL on therapy or residential or MAT is no more restrictive than the parallel NQTL on a medical condition of similar severity. If they cannot, that is your appeal argument.

The “more restrictive than medical” test, in practice

NQTLs are the rules that are not just a number. A 30-visit cap is a quantitative limit. A “you must fail outpatient therapy before we approve IOP” requirement is non-quantitative. MHPAEA cares about both, but the 2024 rule and most of the recent enforcement has been on the NQTL side, because that is where the discrimination tends to hide.

The argument that works in an appeal is structural. You are not just saying “this denial was wrong on the medicine.” You are saying “this denial reflects a process the plan applies more strictly to mental health than to comparable medical care, which violates MHPAEA.”

Concrete scenarios where I have seen this hold up, or at least force the plan to blink:

• Residential mental health or SUD treatment denied as “not medically necessary” while the same plan covers comparable inpatient medical stays (skilled nursing, inpatient rehab after a stroke) without the same level of concurrent review. If the plan reviews residential MH stays every 3 days but reviews inpatient cardiac rehab every 7, that is an NQTL disparity.
• IOP or PHP denied because the plan demands documented failure of outpatient therapy first. Ask whether the plan requires equivalent step-therapy through a lower level of care for, say, cardiac rehab or pulmonary rehab. Usually it does not.
• Therapy session caps applied as a soft denial (“clinically not necessary after 20 visits”) when comparable chronic medical conditions (diabetes management, physical therapy after surgery) are not capped on the same logic.
• Medication for opioid use disorder (buprenorphine, methadone, naltrexone) requiring prior auth, quantity limits, or counseling-attendance requirements that the plan does not impose on comparable chronic disease medications like insulin or statins.
• A behavioral health provider network that, on paper, is half the size of the medical network in the same ZIP code radius. Network composition is itself an NQTL under the rule.

When you frame the appeal, you are layering two arguments. First, the clinical one: the treatment is medically necessary under generally accepted standards (American Society of Addiction Medicine criteria for SUD, the relevant specialty society for MH). Second, the parity one: the plan's denial relies on an NQTL that is more restrictive than the comparable M/S NQTL, in violation of MHPAEA and 45 CFR 146.136.

Plans do not always reverse on parity grounds alone. But the parity argument changes who is reading the file. It pulls the denial out of the medical reviewer's queue and pushes it toward compliance.

How to request the comparative analysis from your plan

This is the move most people do not know they can make. Under the CAA, you (or your authorized representative) can request the plan's NQTL comparative analysis for the specific NQTL that triggered your denial. In writing. The plan is supposed to provide it.

What I usually include in the request:

• Member name, ID, date of denial, the specific service denied.
• The specific NQTL at issue. For example: “the prior authorization requirement for residential SUD treatment,” or “the medical necessity criteria for IOP,” or “the step-therapy requirement before approving IOP.”
• A clear request: “Please provide the comparative analysis required under MHPAEA and the Consolidated Appropriations Act of 2021 for [the specific NQTL], including the factors and evidentiary standards used to design and apply the NQTL, the data used to demonstrate comparability, and the plan's findings and conclusions on compliance.”
• A deadline. I ask for 30 days. The statute is not crystal clear on the exact deadline for participant requests (the 10-business-day clock applies more cleanly to regulator requests), but 30 days is reasonable and gives you something to point to if they stall.

If the plan stalls or sends you a one-paragraph “we comply” letter, that is itself useful. A non-responsive comparative analysis is a finding the DOL has flagged for years. You can include the non-response as part of an external review submission, a DOL complaint (for ERISA group plans), a state DOI complaint (for fully-insured plans), or a CMS complaint (for ACA marketplace and MA-adjacent issues).

I have written more about external review for MA specifically in my IRE request template post. The same logic transfers. Build the file, name the rule, ask the question on the record.

For Medicare Advantage enrollees specifically

This is where I have to be honest with you. Federal MHPAEA, as written, applies primarily to group health plans, ACA individual market plans, and certain Medicaid managed care arrangements. Traditional Medicare is not subject to MHPAEA. Medicare Advantage plans are in an awkward middle place. They are private plans operating under Medicare's coverage rules, governed by 42 CFR Part 422, and they are not directly bound by MHPAEA the way a commercial group plan is.

That does not mean parity arguments are useless against an MA denial. It means you have to translate them.

The translation paths I use:

• Medicare coverage rules. MA plans must cover everything Original Medicare covers. If Medicare covers a level of mental health or SUD care under its national or local coverage determinations, the MA plan cannot apply a more restrictive rule and call it medical management. Cite the relevant Medicare coverage policy.
• CMS network adequacy. Under 42 CFR 422.116, MA plans have to meet time-and-distance and minimum-number standards for provider access. For 2026, CMS added outpatient behavioral health as a new facility-specialty provider category that counts toward network adequacy. If you got denied because there was no in-network provider, or because the in-network provider had no availability for 8 weeks, that is a network adequacy complaint to CMS, separate from the appeal.
• MA grievance process. Appeals address a specific coverage decision. Grievances address how the plan operates. Network gaps, repeated parity-pattern denials, and unresponsive plans go in grievances. File both.
• State Department of Insurance. Even though MA is federally regulated, many states will take complaints about an MA plan's behavior, especially if the same insurer also writes commercial business in the state. The state DOI will often forward the complaint to CMS, but the file gets built either way.
• State parity laws. If you live in a state with a strong parity statute (more on that below), and the MA plan is operated by an insurer that is also licensed in the state for commercial business, state DOI complaints can still apply pressure even where federal MHPAEA does not directly govern.

The mechanics of the MA appeal itself are unchanged. You have 65 days from the date on the denial notice to file a Level 1 reconsideration with the plan. If the plan upholds, it forwards automatically to the Independent Review Entity for Level 2. As of May 1, 2026, C2C Innovative Solutions is the Part C IRE. I walk through the IRE step in detail in my external review template post, and through the broader MA appeal mechanics in the prior-authorization denial post.

The other piece I want to flag for MA enrollees: a lot of mental health and SUD denials at the MA level look exactly like the post-acute denials I wrote about in the rehab-cut-short post. Same algorithmic-feeling rationale. Same vague “no longer medically necessary.” The appeal craft is similar. Build the file. Name the rule. Force a human review.

State-level parity, and when it beats federal

Federal MHPAEA is the floor, not the ceiling. Several states have parity laws that are stronger than the federal framework, and the state DOI is often the faster and more responsive enforcement venue for an individual case, especially for fully-insured plans (self-funded ERISA plans are still federal-only).

A non-exhaustive list of states whose parity statutes go beyond federal MHPAEA in at least some respects: California, New York, Illinois, Massachusetts, Oregon, Washington, Colorado, Connecticut, Maryland, Pennsylvania, Delaware, New Mexico, Tennessee. New York and Illinois in particular have taken public enforcement actions against major carriers for parity violations in the last few years. California, New Mexico, and Oregon require plans to use generally accepted standards of care (often tied to specialty society criteria) for medical necessity decisions, which closes one of the favorite escape hatches plans use.

How I think about venue choice:

• Self-funded employer plan, anywhere in the country: federal (DOL EBSA) is your enforcement path. State DOI cannot help.
• Fully-insured employer plan or individual market plan in a strong-parity state: file at the state DOI first. They tend to move faster and they care about the carrier's license.
• Fully-insured plan in a weaker-parity state: federal DOL or HHS, depending on plan type.
• ACA marketplace plan: HHS / CMS, plus state DOI if the state has its own marketplace.
• Medicare Advantage: CMS, plus state DOI as a secondary pressure point if the carrier also writes commercial in the state.

I keep a running file on state DOI complaint processes on my own site, but for parity specifically, the state department's consumer services line is usually the fastest first call. They will tell you what they accept.

Practical framing for the appeal letter itself

You do not need to write a legal brief. You need to put the parity argument on the record clearly enough that the next reviewer cannot miss it. Something like this, adapted to your facts:

The plan denied coverage for [residential SUD treatment / IOP / 30 additional therapy sessions / buprenorphine prior authorization] on [date] under [the specific medical necessity criterion or step-therapy rule the plan cited]. This denial relies on a non-quantitative treatment limitation that, on its face and in operation, is more restrictive than the comparable NQTL the plan applies to medical/surgical benefits of similar severity. Under MHPAEA and 45 CFR 146.136, an NQTL applied to MH/SUD benefits must be comparable to, and applied no more stringently than, the NQTL applied to comparable M/S benefits. Pursuant to the Consolidated Appropriations Act of 2021, I am also requesting the plan's written NQTL comparative analysis for [the specific NQTL at issue], including the factors and evidentiary standards used, the data evaluated, and the plan's findings on compliance. Please provide the comparative analysis within 30 days.

Attach the clinical record. Attach the provider's letter of medical necessity if you have one. If you can name a comparable medical condition the plan covers without the same restriction, name it.

That paragraph is not magic. It is just a way to make the parity issue impossible to ignore.

If you want this filed for you

I run DenyBack. It is $39, flat, one denial. I write the appeal letter, build the parity argument, draft the comparative-analysis request, and prep the state DOI or CMS complaint if it makes sense. I do not take a cut of anything the plan pays. I do not upsell. The $39 is the whole thing.

I do this because I think single-appeal pricing is the only honest model for someone facing a denial in the middle of a mental health or addiction crisis. Contingency fees do not really work here. Hourly legal help is out of reach for most people. A flat fee for one piece of writing that gets the rule on the record is what I can offer.

If you want to use DenyBack, the site is denyback.com. Send me the denial letter and the relevant clinical notes. I will turn the appeal around in a few business days.

Permission to ignore the pitch

If you do not want to use DenyBack, please do not. Use this post. Use the template paragraph. Request the comparative analysis yourself. File at your state DOI. The point is that the appeal gets written, not that I write it. The plans count on people not writing back. The single most important thing you can do is write back, name the rule, and put the file on the record.

If you do that and the denial sticks, you still have the external review or IRE step, and the parity argument carries forward into that step. None of this is wasted work.