Mental Health Parity in Medicare

Medicare has never been subject to the Mental Health Parity and Addiction Equity Act. The Mental Health Parity Act of 1996 and the MHPAEA of 2008, which required private health plans to cover mental health and substance use disorders on terms no more restrictive than coverage for medical and surgical conditions, were explicitly written not to apply to Medicare. In 2016, parity rules were extended to Medicaid managed care organizations but, again, not to Medicare benefits provided by those same organizations to dual eligibles. The result is that the federal program covering more than 67 million Americans, including most people with serious mental illness who are old enough or disabled enough to qualify, operates outside the core legal framework that governs how every other form of federally regulated insurance must treat behavioral health.

This is not a gap that regulatory interpretation or Stars incentives fully resolve. The structural disparities are embedded in statute, and closing them requires either congressional action or a sustained administrative effort to use existing authorities in ways that move closer to parity outcomes without legislating them.

The Historical Structural Problem

#

The most visible statutory artifact is the 190-day lifetime limit on inpatient care in freestanding psychiatric facilities. Medicare imposes no equivalent lifetime limit on any other category of inpatient specialty care. A beneficiary with cancer can be hospitalized repeatedly for chemotherapy, a beneficiary with heart failure can require multiple cardiac admissions, and neither faces a cap on covered lifetime days specific to their condition. A beneficiary who requires recurring inpatient psychiatric stabilization for schizophrenia, bipolar disorder, or major depressive disorder faces a hard ceiling: 190 days, total, across their entire Medicare enrollment history, in facilities that specialize in treating what they have.

As of January 2024, approximately 40,000 Medicare beneficiaries had exhausted the 190-day limit. Another 10,000 were within 15 days of it. Among those near or at the limit, over 70 percent were under 65, meaning they qualified for Medicare through disability rather than age. Eighty percent had a diagnosis of schizophrenia in the prior year. Thirty-four percent had a co-occurring substance use disorder. These are, systematically, the people who need the most psychiatric inpatient care: younger, severely mentally ill, often low-income, often dually eligible.

The 190-day limit applies only to freestanding inpatient psychiatric facilities. Psychiatric units housed within general acute care hospitals are governed by standard Medicare Part A inpatient rules, with no equivalent lifetime cap. This creates a structural inequity within the psychiatric inpatient system itself: patients treated at specialized psychiatric hospitals face a lifetime ceiling that patients treated in general hospital psychiatric units do not. There is no clinical rationale for the distinction. It reflects the historical structure of Medicare’s benefit design, not any evidence about where intensive psychiatric care produces better or worse outcomes.

Before the Affordable Care Act, Medicare charged 50 percent coinsurance for outpatient mental health services, compared to 20 percent for other Part B services. The ACA phased that differential down over five years, reaching parity at 20 percent coinsurance by 2014. That reform addressed the most visible cost-sharing disparity in the outpatient setting, but it did not extend MHPAEA to Medicare, and it did not address the 190-day cap, the absence of residential SUD treatment coverage, or the utilization management practices in Medicare Advantage that operate in a legal environment where parity enforcement does not apply.

The Provider Access Disparity

#

The coverage structure is one layer of the parity gap. The provider access layer is the other, and it is addressed in detail in the first two articles of this series. The short version: Original Medicare covers psychiatric evaluations, psychotherapy, medication management, partial hospitalization, intensive outpatient services, and collaborative care billing codes. It pays psychiatrists, clinical psychologists, licensed clinical social workers, and, since January 2024, marriage and family therapists and mental health counselors. The CAA 2023 expansion to MFTs and MHCs added an estimated 400,000 potentially eligible providers to the Medicare-billable workforce, a meaningful expansion that has not yet translated into measurable access improvement because participation rates remain low and Medicare’s reimbursement rates for behavioral health services create weak financial incentives for private practitioners to join.

The ghost network problem in MA is well-documented. OIG’s October 2025 report on behavioral health provider networks found that 55 percent of providers listed in MA plan behavioral health directories were inactive in the studied counties. Fifteen of the 40 MA plans studied had networks that included fewer than 10 percent of the available behavioral health workforce in the relevant counties. Seven MA plans had no in-network behavioral health providers in the studied counties at all. A 2023 Senate Finance Committee investigation found that when staff called providers listed as in-network with MA plans for mental health services, more than 80 percent were either unreachable, not accepting new patients, or not actually in-network.

The time-distance standards that CMS uses for MA network adequacy have not resolved the ghost network problem. An MA plan can technically satisfy CMS network adequacy requirements while maintaining a directory full of providers who are not taking new patients, have left the practice, or stopped participating in Medicare. CMS audits have not targeted behavioral health specifically, which is the finding at the center of the May 2025 GAO report on MA prior authorization and behavioral health. GAO found that 8 of 9 selected MA organizations required prior authorization for behavioral health services, and that CMS oversight of the prior authorization criteria used for behavioral health had not assessed effects on enrollees’ access to care.

The Prior Authorization Disparity

#

The GAO’s May 2025 report, covering nine MA organizations that together served about 45 percent of MA beneficiaries in 2024, found that prior authorization requirements applied to a range of behavioral health services including inpatient psychiatric care, partial hospitalization, intensive outpatient programs, and, in some organizations, outpatient therapy. The report specifically found that CMS had not released final plans for auditing whether prior authorization criteria for behavioral health services meet Medicare coverage rules, a gap that creates de facto enforcement immunity for utilization management practices that may not be consistent with covered benefits.

This matters because the utilization management asymmetry between behavioral health and physical health is a documented MHPAEA compliance problem in commercial insurance. Plans subject to MHPAEA are prohibited from applying prior authorization requirements to mental health services that are more burdensome than requirements for comparable medical and surgical services. Because MHPAEA does not apply to Medicare, MA plans can apply prior authorization to behavioral health services in ways that would be legally problematic under commercial parity rules, without CMS audit specifically targeting those practices.

The structural logic is straightforward: without a parity obligation, without targeted enforcement, and in a context where the behavioral health provider supply problem means that many services that are authorized are never actually delivered due to network gaps, the disincentive to seek behavioral health care compounds at every step. A beneficiary who identifies a covered service, finds a participating provider, obtains a referral, and then faces prior authorization requirements is navigating a system with more friction than the equivalent process for most physical health services.

HIDE SNP as Partial Parity

#

The HIDE SNP framework, described in detail in MCR-08.02, creates behavioral health integration requirements for plans serving dual eligible populations specifically. A HIDE SNP that qualifies as an Applicable Integrated Plan must have a capitated Medicaid contract that covers behavioral health services, long-term services and supports, or both. FIDE SNPs are required to cover behavioral health starting in 2025, subject to state carve-out limitations. HIDE SNPs must screen for social determinants of health, including behavioral health-relevant domains, as a condition of their D-SNP designation.

For the dual eligible population specifically, these requirements move the coverage structure closer to parity than standard MA. A beneficiary enrolled in a FIDE or HIDE SNP in a state where behavioral health is not carved out of Medicaid managed care has both Medicare behavioral health coverage and Medicaid behavioral health coverage, potentially including services Original Medicare does not cover: Assertive Community Treatment, crisis stabilization, residential treatment, peer support. The integrated plan structure creates a vehicle for care coordination between these streams that does not exist for beneficiaries in standard MA or Original Medicare.

HIDE does not, however, apply parity principles to standard MA plans or to Original Medicare. It is a targeted coverage expansion for a subset of dual eligibles in states and counties where FIDE or HIDE SNPs operate, which as of 2024 excluded more than a third of D-SNP enrollees by county availability. The compliance-aspiration gap documented in MCR-08.02 applies here as well: HIDE SNP behavioral health integration requirements create a mandate, but without adequate provider supply and without enforcement infrastructure that tracks actual service delivery rather than plan design attestations, the mandate does not automatically produce integrated care.

The MHPAEA Enforcement Question

#

CMS’s regulatory authority over MA behavioral health operates through several overlapping frameworks. Network adequacy standards require MA plans to maintain sufficient numbers of behavioral health providers to meet time-distance and ratio standards. Non-discrimination provisions in the MA statute prohibit plans from discriminating against beneficiaries based on health status. CMS has authority under the MA regulatory structure to review prior authorization criteria for consistency with Medicare coverage rules, which is the authority the GAO identified as not being targeted at behavioral health.

Whether CMS has authority to require formal MHPAEA compliance in MA without new legislation is a contested legal question. MHPAEA applies to group health plans and health insurance issuers in the group and individual markets, Medicaid managed care, and CHIP. MA plans are structured as Medicare contracts rather than health insurance issuers in the MHPAEA definitional sense, and the statute does not extend the parity obligation to them. CMS has never formally analyzed whether existing MA statutory authority is sufficient to impose MHPAEA-equivalent requirements administratively.

What CMS can do without new legislation is use existing authorities more aggressively. Targeted behavioral health audits of prior authorization criteria, as recommended by GAO, would identify whether MA plans are applying coverage criteria that are inconsistent with Medicare coverage rules. Enforcement of network adequacy requirements against plans with documented ghost networks in behavioral health would address the most visible access gap. Requiring MA plans to report behavioral health-specific utilization management data, including prior authorization approval and denial rates by service category, would create the transparency necessary to identify patterns of de facto coverage restriction.

The 2024 final rule for contract year 2026 included provisions addressing prior authorization transparency and required plans to specify clinical criteria for prior authorization decisions. The final rule issued in April 2025 deferred the provisions on internal coverage criteria for subsequent rulemaking. This deferral means that the framework CMS was developing to ensure that MA organizations apply prior authorization criteria consistent with Medicare coverage rules remains incomplete for behavioral health services specifically.

The Legislative Landscape

#

H.R. 4619, the Medicare Mental Health Inpatient Equity Act of 2025, was introduced in the 119th Congress to permanently repeal the 190-day lifetime limit on inpatient psychiatric care. The bill has bipartisan support and advocacy from NAMI, the American Psychiatric Association, and mental health advocacy organizations. The MedPAC March 2025 report to Congress included a chapter recommending elimination of both the 190-day lifetime limit and the associated benefit days reduction formula, citing the lack of clinical rationale and the disproportionate impact on younger, severely mentally ill, low-income beneficiaries.

The CBO score for eliminating the 190-day cap is bounded. The population actively affected, approximately 50,000 beneficiaries at or near the limit in any given year, is relatively small. MedPAC’s analysis indicates that full-benefit dual eligibles over 65 who reach the Medicare cap often have Medicaid coverage for additional days, meaning the marginal Medicare cost of repeal for that subset is lower than the headline figure suggests. The harder scoring question is the behavioral response: repeal of the cap might induce additional inpatient psychiatric utilization from beneficiaries who currently self-ration care because they are managing their lifetime days. Any CBO estimate would need to model that behavioral effect.

Applying MHPAEA to Medicare in full is a larger and more expensive proposal. It would require Congress to amend MHPAEA to include Medicare, direct CMS to conduct comparative analyses of coverage limitations for behavioral health and medical-surgical services across all Medicare benefit categories, and potentially redesign several statutory coverage limitations that would be classified as non-compliant quantitative treatment limitations under the parity framework. The 190-day cap is the most obvious, but the absence of residential SUD treatment coverage, the absence of non-hospital psychiatric residential coverage, and the reimbursement differential for behavioral health providers would all be subject to review.

The incremental administrative pathway, using Stars, network adequacy standards, and HIDE requirements together, can move the access picture without resolving the statutory parity gap. The proposed depression screening Stars measure creates quality incentive alignment for MA plans to build systematic behavioral health identification into primary care. Network adequacy enforcement against ghost networks would force plans to either contract with providers who are actually available or acknowledge that their networks are deficient. HIDE and FIDE SNP requirements are extending integrated behavioral health to the highest-need dual eligible populations as the D-SNP market matures.

These tools do not constitute parity. They create pressure in the direction of parity for specific populations in specific coverage contexts. Full Medicare parity requires Congress to decide whether the program it built in 1965, and has amended more than three hundred times since, should finally be required to cover the minds of its beneficiaries on the same terms as their bodies.

Related Reading

#

MCR-03_03 Medicare Equity: What the HEI Reversal Signals and What Remains MCR-10_03 LGBTQ+ Seniors in Medicare: Coverage Gaps, Plan Discretion, and the Non-Discrimination Framework