Network Adequacy and Access

A plan can cover behavioral health on identical terms to medical care and still deliver far less of it, if its behavioral network is thin, its directories are inaccurate, or its rates are low enough that clinicians decline to participate. This is the access problem that formal coverage parity does not by itself solve.

Why behavioral networks are thin

The mechanism is economic rather than mysterious. Behavioral clinicians, particularly psychiatrists and psychologists in areas with sufficient demand, can fill a practice with patients paying directly. Joining a network means accepting negotiated rates and administrative burden in exchange for volume the clinician does not need.

The result is a network that exists on paper and performs poorly in practice: listed clinicians not accepting new patients, no longer at the listed location, or not practicing at all. A directory entry is not access.

Parity treats this as a limitation

Federal regulation now recognizes this explicitly. Among the examples of nonquantitative treatment limitations subject to parity requirements are “standards related to network composition, and methodologies to determine out-of-network reimbursement rates.”

That framing is the substantive step. It means a plan cannot satisfy parity merely by matching copays and visit limits; how it builds its network and what it pays out-of-network providers are themselves subject to comparison against the medical side.

The 2024 final rules went further, requiring plans and issuers “to collect and evaluate data and take reasonable action, as necessary, to address material differences in access to MH/SUD benefits as compared to medical/surgical benefits that result from application of NQTLs, where the relevant data suggest that the NQTL contributes to material differences in access.”

That is an outcomes test rather than a design test: not whether the rule looks equal, but whether access is. Module 4 covers what happened to those provisions.

What accountable organizations can do

For a provider organization at risk, network thinness is not an abstraction, it is why referrals fail. The available responses are the ones this curriculum has covered elsewhere, applied here:

• Build capacity rather than referring into scarcity. Collaborative care exists because referral capacity does not.
• Contract directly with behavioral groups on terms that make participation worthwhile, which usually means better rates or guaranteed volume.
• Measure referral completion, not referral generation. A referral that never becomes a visit is a failure the organization currently cannot see.
• Use telehealth deliberately, since behavioral care is among the services where remote delivery is closest to equivalent and geography is the binding constraint.

Worth remembering: access problems in behavioral health are frequently described as workforce shortages, which is true and incomplete. A shortage explains why some patients wait. It does not by itself explain why clinicians who exist decline to join networks, and that second question has an answer involving rates and administrative burden that a plan controls. Treating the whole problem as a shortage puts it outside anyone’s responsibility. Separating the part that is genuinely scarcity from the part that is pricing is what makes it actionable.

Key takeaways

• Coverage parity does not guarantee access; thin networks and inaccurate directories limit it.
• Federal parity regulation names network composition and out-of-network reimbursement as nonquantitative treatment limitations.
• The 2024 rules required evaluating whether limitations produce material differences in actual access.
• Provider organizations can build capacity, contract directly, measure referral completion, and use telehealth.

Sources

• CMS, The Mental Health Parity and Addiction Equity Act (MHPAEA) (opens in a new tab)
• KFF, 5 Key Facts About Medicaid Coverage for Adults with Mental Illness (opens in a new tab)