The Laws That Were Supposed to Protect Us

Part 3 of 5 in the series Connecting the Dots: The Price of Fragmentation in Mental Healthcare

MHPAEA, ERISA, and the gap between what the law promises and what clinicians and clients actually experience

When the Mental Health Parity and Addiction Equity Act (MHPAEA) passed in 2008, it was celebrated as a landmark. Mental health care would finally be treated like physical health care. Insurance companies would no longer be able to impose more restrictive limits on behavioral health benefits than on medical and surgical benefits. After decades of discrimination baked into coverage structures, federal law was drawing a line.

The line it drew was much narrower than almost anyone understood at the time. And the largest category of private health coverage in the country sits almost entirely outside it.

I want to be precise about this, because the gap between what parity law promises and what it actually delivers is one of the most important and least understood problems in the entire behavioral health system. Understanding it is essential to understanding why advocacy efforts that focus on state-level reform keep hitting a wall.

What MHPAEA Actually Covers

The Mental Health Parity and Addiction Equity Act (MHPAEA) prohibits group health plans from imposing more restrictive financial requirements or treatment limitations on mental health and substance use disorder benefits than on medical and surgical benefits. Specifically, it prohibits higher copays, higher deductibles, stricter visit limits, and more restrictive prior authorization requirements for behavioral health than for comparable medical services.

That is a genuine protection. Before parity laws, it was routine for insurance plans to charge a $50 copay for a therapy session and a $20 copay for a primary care visit, or to cap mental health visits at 20 per year while placing no limits on medical visits. MHPAEA ended those specific practices for the plans it covers.

But here is what it does not cover, stated plainly: MHPAEA does not regulate what insurance companies pay providers. The financial requirements it prohibits are the patient’s costs at point of service: copays, deductibles, out-of-pocket maximums. The reimbursement rate an insurer pays a therapist is a transaction between the insurer and the provider. That transaction is entirely outside the scope of the law.

This is not an oversight or a technicality. It is the central structural limitation of the entire parity framework. An insurer can charge a patient the exact same copay to see a therapist as to see their cardiologist (parity achieved on paper), while simultaneously reimbursing that therapist at a rate so low that no therapist can afford to participate in the network. The patient’s copay is equal. The provider’s reimbursement makes the network functionally inaccessible. The law protected the visible inequity while leaving the structural economic inequity that drives network inadequacy completely intact.

The 2024 final rules did begin to move toward reimbursement through the Non-quantitative Treatment Limitation (NQTL) framework. Agency guidance has stated that a methodology that reimburses in-network behavioral health providers at the Medicare rate while reimbursing medical health providers at two times the Medicare rate violates parity law. The agencies have also identified as impermissible disparities in the method used to determine reimbursement for non-physician mental health practitioners compared to non-physician medical practitioners. This is a meaningful development, but it targets reimbursement methodology, not the resulting rate. An insurer can use the same methodology and still produce a lower rate if the underlying inputs differ. And enforcement of even this narrower protection has been minimal.

The ERISA Exemption: Where Parity Goes to Die

When the Mental Health Parity and Addiction Equity Act (MHPAEA) passed in 2008, it was celebrated as a landmark. Mental health care would finally be treated like physical health care. Insurance companies would no longer be able to impose more restrictive limits on behavioral health benefits than on medical and surgical benefits. After decades of discrimination baked into coverage structures, federal law was drawing a line.

The line it drew was much narrower than almost anyone understood at the time. And the largest category of private health coverage in the country sits almost entirely outside it.

Here is the part that almost nobody outside healthcare policy circles understands, which has profound consequences for whether any of the above matters in practice:

Approximately 64% of Americans with employer-sponsored health insurance are covered by self-funded plans. When an employer self-funds its health plan, meaning it pays claims out of its own assets rather than buying a policy from a carrier, that plan is not legally "insurance" at all. Under ERISA's deemer clause, a self-funded employee benefit plan cannot be deemed an insurance company, an insurer, or to be engaged in the business of insurance for purposes of any state law regulating insurance. It is an employee benefit governed by federal law, not an insurance product. That means the state insurance commissioner has no jurisdiction over it, and the state benefit mandates, network rules, and consumer protections that apply to commercial insurance simply do not reach it. The plan looks like insurance to the patient and is often administered by a familiar carrier acting as a third-party administrator, but the carrier is processing claims, not insuring the risk, and the legal protections people assume they have are not there.

The Employee Contribution Problem

ERISA preemption was designed to allow multistate employers to administer uniform benefit plans without navigating 50 different state regulatory regimes. The stated rationale is protecting employers’ ability to offer benefits efficiently. That rationale deserves scrutiny when you look at how these plans are actually funded.

When an employee contributes 50% of their premium or pays 100% of their dependent coverage out of pocket, which is increasingly common as employers shift costs, the employer-as-plan-sponsor relationship that shields the plan from state regulation becomes increasingly difficult to defend on its own terms. An employee paying $800 a month for their family’s coverage through an employer plan that happens to be self-funded is not the beneficiary of an employer’s voluntary generosity. They are a consumer purchasing a financial product with their own money, routed through an employer for tax reasons. The legal fiction that this transaction should be governed by a 1974 pension protection law rather than state insurance consumer protection law because the employer’s name is technically on the plan produces real harm to real people who have no idea they lack the protections they think they have.

Dependent coverage is the sharpest version of this argument:

When an employee pays 100% of the premium for their spouse and children; no employer contribution, employer serving purely as administrative conduit, there is no coherent policy rationale for why that family’s mental health claims should be adjudicated without state consumer protection, bad faith liability, or network adequacy oversight. That is a consumer buying a health product. The product happens to be routed through an employer. The ERISA shield protecting it from state regulation is protecting no legitimate employer/employee interest at that point.

This is an argument that has not been made clearly or loudly in the current legislative conversation. It is a specific, bounded, principled argument for conditioning ERISA preemption on meaningful employer contribution. If the employee bears more than a defined threshold of total premium cost, state consumer protection and insurance laws should apply. That is not a general attack on ERISA preemption, which large employer lobbying groups defend ferociously. It is a targeted reform with a defensible rationale that addresses the most egregious version of the problem.

The Combined Effect

When you put MHPAEA and ERISA together, the picture is this: the federal parity law that was supposed to end mental health coverage discrimination explicitly does not cover provider reimbursement rates, and applies to a minority of the market because the majority is ERISA-preempted from state law. The stronger state parity protections that fill some of the gaps in federal law cannot reach the 64% of employer-sponsored coverage that operates under ERISA. And the enforcement of even the federal protections that do apply is chronically under-resourced.

The result is a system that looks like parity from a distance and functions as anything but up close. Clinicians experience it every day; rates that do not reflect the complexity of the work, ghost networks that list them as accepting new patients when they are not, prior authorization processes for behavioral health that have no equivalent in medical benefits, and carve-out structures that create separate administrative labyrinths for mental health claims. Much of this is legal under current law. Some of it violates existing law and goes unenforced. Almost none of it is visible to the patients trying to navigate it.

The profession’s failure to build unified advocacy infrastructure did not just cost us Medicare recognition and reimbursement rates. It cost us the organized presence and political capital to shape these legal frameworks as they were being built. MHPAEA was passed without a provision explicitly covering provider reimbursement rates because no unified behavioral health workforce organization was at the table with the standing and the resources to fight for it. ERISA has preempted state mental health protections for fifty years because no behavioral health coalition has mounted the sustained legislative campaign to carve out an exception.

These are not failures of law. They are failures of advocacy. And they are failures we could begin to address if the profession ever decides to organize around what clinicians actually need rather than around what makes each guild feel distinct.

Part Four examines what happened inside the training programs while all of this was unfolding outside them and how it set the stage for the commodification of our field.