Access & Policy

Network Adequacy Failures in Mental Health: What Health Plans Can Do Now

Map with sparse provider location pins representing mental health network adequacy gaps

Mental health network adequacy is no longer an abstract compliance problem. It is a documented enforcement target. The Centers for Medicare & Medicaid Services finalized network adequacy standards for Marketplace plans that explicitly apply quantitative time-and-distance requirements to mental health providers. State insurance commissioners are conducting surprise audits of provider directory accuracy — the so-called "ghost network" problem, where listed providers are not accepting new patients. And class action litigation has named health plans in suits alleging that on-paper network parity with medical/surgical coverage conceals real access disparities.

Health plan medical directors and network operations teams operating in this environment face a genuine structural problem: the behavioral health workforce shortage is not a short-term capacity constraint. HRSA projections consistently show a psychiatrist and psychologist shortage that extends into the 2030s. You cannot contract your way to network adequacy if the providers don't exist in the relevant geographies. This post is about what plans can actually do within that constraint — not aspirational workforce development, but operational options available now.

Understanding What Network Adequacy Standards Actually Require

The Mental Health Parity and Addiction Equity Act (MHPAEA) of 2008, and its implementing regulations, require that quantitative treatment limits (visit limits, prior auth rates) and non-quantitative treatment limits (NQTLs) for mental health and substance use disorder benefits be no more restrictive than the predominant standards applied to medical/surgical benefits. Network composition and provider access are NQTLs — which means plans must analyze whether their behavioral health networks impose more restrictive access conditions than their medical/surgical networks when measured against the same metrics.

The 2024 MHPAEA final rule tightened this requirement by mandating comparative analyses of NQTLs in writing, available to regulators on request. Plans that cannot document that their behavioral health network adequacy standards are applied comparably to medical/surgical are exposed. The days of relying on implicit parity are over.

CMS network adequacy standards for QHP plans add time-and-distance requirements: for outpatient behavioral health, the 2024 standards set maximum travel time and distance thresholds by geography type (urban, suburban, rural) within which a specified percentage of members must have access to in-network providers. The specific thresholds vary by state and plan type, but the direction of travel is toward quantitative measurability rather than self-attestation.

The Ghost Network Problem Is a Compliance Problem, Not Just a Member Experience Problem

Multiple state insurance department audits in 2022-2024 have found that between 40% and 65% of listed in-network behavioral health providers are not accepting new patients when contacted. This is the ghost network problem: a provider directory that passes adequacy screening on paper but fails in practice.

State regulators increasingly treat directory inaccuracy as an independent compliance failure distinct from network adequacy. California, New York, and Texas have all issued guidance or enforcement actions specifically targeting provider directory accuracy for behavioral health. The consequence for plans is that directory remediation — not just network expansion — is now a compliance task that requires operational infrastructure.

Plans that have not conducted a recent audit of their behavioral health provider directory against real appointment availability are operating with regulatory exposure they may not have quantified. The audit methodology matters: secret shopper calls to a random sample of listed providers, tracked against appointment availability and reported wait time, is the standard approach. Plans should be doing this quarterly, not annually.

Telehealth Expansion Is Necessary but Not Sufficient

The behavioral health telehealth expansion that followed 2020 public health emergency provisions has become a permanent structural feature of mental health care delivery for many plans. Telehealth meaningfully expands effective network reach for members who have internet access and are comfortable with video-based care. For rural and suburban geographies where in-person behavioral health access is most constrained, telehealth providers have measurably improved the match between listed and available providers.

But telehealth has real coverage limits. Approximately 22% of rural Americans lack reliable broadband access (FCC 2022 data). Members experiencing moderate-to-severe depression or anxiety may find telehealth sessions harder to engage with than in-person care — a smaller but real clinical population. And telehealth expansion does not resolve the psychiatry shortage for medication management, which is not adequately addressed by licensed clinical social workers or marriage and family therapists regardless of delivery modality.

We're not saying telehealth network expansion is inadequate — it is a necessary component of access strategy. The point is that it doesn't exhaust the plan's obligation or the member's need.

Digital Therapeutics as a Network Adequacy Complement

The category of digital mental health tools — apps and conversational AI platforms delivering structured CBT, DBT skills training, or mindfulness-based intervention — has matured to the point where some products have clinical evidence supporting their use as a legitimate complement to limited network access. This is different from saying digital tools replace therapy; they don't. But for a member with a 6-week wait for an outpatient therapist, structured digital CBT during the wait period has documented benefit.

CMS has not yet formally recognized digital therapeutics as network adequacy supplements in quantitative terms, but state-level regulators in states with progressive digital health frameworks (California, Washington, Massachusetts) have begun including app-based mental health support in network access discussions. MHPAEA compliance analysis that includes digital tool availability as part of the access picture is still emerging, but the direction of policy travel supports including it in access calculations for lower-acuity presentations.

What Neurodex offers health plans in this context is step-down CBT support — structured, trajectory-monitored sessions for members whose clinical acuity is appropriate for structured self-guided therapy with AI support, with explicit escalation protocols for members whose symptom severity warrants human clinical review. The PMPM cost of this tier is a fraction of the cost of even a single outpatient session, making it viable as a standard benefit for the large population of members with mild-to-moderate presentations who are currently getting nothing while they wait.

What Plans Can Do in the Next 90 Days

Practical steps available to health plan operations teams without long-term vendor commitments or regulatory approval cycles:

Provider directory audit. Commission or conduct a secret shopper audit of your behavioral health provider directory. Target a random sample of 50-100 listed providers. Track appointment availability, new patient acceptance, and reported wait time. Compare against your medical/surgical directory for the same metrics. The MHPAEA NQTL analysis requires this comparison; having the data before a regulator asks for it is strongly preferable to reconstructing it under audit conditions.

Telehealth network gap analysis. Map your telehealth behavioral health provider coverage against member geography. Identify concentrations of members in geographies where telehealth access is below your standard adequacy thresholds, either because of broadband limitations or because your telehealth network doesn't extend to those zip codes. These are your highest-priority network gaps.

NQTL comparative analysis documentation. If your MHPAEA NQTL analysis has not been updated since the 2024 final rule, update it. Specifically, document the criteria used to determine behavioral health network adequacy and confirm those criteria are at least as favorable as the criteria used for medical/surgical network adequacy. Legal and compliance exposure from an undocumented or outdated NQTL analysis has materially increased since 2024.

Pilot a digital CBT benefit for waitlisted members. For members who have been triaged for outpatient behavioral health and are waiting for an appointment, a structured digital CBT benefit during the wait period has evidence support and minimal implementation complexity. This addresses both access and outcomes — members who arrive at their first in-person session having already done structured work on automatic thoughts and activity scheduling have better clinical trajectories than those who waited passively.

The Uncomfortable Structural Reality

The behavioral health provider shortage is a structural feature of the US healthcare labor market that no health plan can unilaterally solve. The shortage of psychiatrists in rural areas is not going to be resolved by improved contracting practices. Plans that face network adequacy enforcement are, in many cases, being held to standards that the provider workforce cannot satisfy — a legitimately unfair situation that nonetheless creates real legal and regulatory risk.

The practical posture for network operations leaders is to document, analyze, and demonstrate good-faith effort to maximize access within the available workforce. That means accurate directories, strong telehealth networks, digital adjuncts for appropriate clinical populations, and written NQTL analyses that don't require reconstruction under audit pressure. None of this fully solves the access problem. All of it meaningfully reduces compliance exposure and, more importantly, gets more members into structured care faster.

Interested in deploying Neurodex for your health plan or EAP? Request a pilot.

Request a Pilot