Ghost networks get a price tag in Carelon's $10M New York deal
A reported $10 million settlement over inaccurate provider listings and coverage denials turns directory data integrity into a measurable compliance liability for payer executives heading into 2027.

The ghost networks payer settlement reported this week puts a dollar figure on a problem health plans have treated as a back-office nuisance for years. Carelon agreed to pay $10 million to resolve New York allegations involving inaccurate provider listings and improper coverage denials, according to Fierce Healthcare. For payer executives, the number matters less than the precedent: directory data quality is now priced.
Two failures, one enforcement action
What makes the Carelon matter instructive is that it bundles problems most plans manage in separate silos. The allegations cover inaccurate or unreachable provider listings, which is a data and operations issue, alongside improper denial of coverage, which sits with utilization management and clinical policy. Regulators treated them as a single access failure rather than two unrelated compliance lapses.
That framing has consequences. A plan can pass a network adequacy filing on paper while members cannot actually reach a behavioral health clinician, then compound the problem when the care that does get delivered out of network gets denied. Enforcement bodies have been building this argument for roughly two years in behavioral health. A named subsidiary of a major national insurer and an eight-figure figure is the escalation point.
Payer leaders should note the reporting caveat. Public detail on the named entity, the statutes cited and any corrective action terms remains thin. Legal and compliance teams reading this should confirm the specifics before using the case internally as a benchmark.
Automating on top of unreliable network data does not neutralize the exposure. It scales it.
Why directory accuracy is harder than it sounds
Provider directories are downstream of credentialing systems, roster feeds from delegated groups, contract management platforms and provider self-attestation. Each of those inputs has its own refresh cadence, its own owner and its own error rate. The public-facing directory is where all of that drift becomes visible to a member making a phone call.
Behavioral health is the hardest case. Panels are heavily solo and small-group, clinicians move or close panels without notifying plans, and acceptance of new patients changes faster than any attestation cycle captures. A listing can be technically accurate on licensure and address while being functionally useless to a member seeking an appointment. Regulators have increasingly adopted the member's definition of accuracy, not the plan's.
For health plan CIOs, the operational implications are concrete: shorten attestation cadence, instrument reconciliation between credentialing source of truth and the consumer directory, and build secret-shopper style validation into routine operations rather than treating it as an audit response.
The AI spending paradox
The settlement lands in the same news cycle as CareSource's new $1 million grant program for AI-powered solutions, part of a broader pattern of plans directing discretionary dollars toward automation while trimming administrative cost. Fierce Healthcare's third-quarter recap of payer trends documents continuing margin and utilization pressure heading into 2027, which explains the appetite for tooling that promises to do more with fewer staff.
Here is the tension worth owning. AI tools for network management, member navigation and care routing read the same roster and directory data that regulators are now fining plans over. Automating on top of unreliable network data does not neutralize the exposure. It scales it, and it makes the error path harder to audit after the fact.
The sequencing argument for boards is straightforward: data integrity investment is not a competing priority with AI investment. It is the precondition for it.
What health system and behavioral health leaders should do with this
On the provider side, the settlement is negotiating leverage. If a plan's network looks adequate only because the directory says so, both contracted rates and access obligations are open questions at the table. Behavioral health groups in particular have long argued that low participation reflects rates rather than clinician supply. An enforcement action that puts a price on paper networks strengthens that case.
Practical steps: document directory errors involving your own clinicians, including closed panels listed as open and clinicians who left the organization but remain listed. That record is useful in contract renewals and in any state regulatory complaint. Track referral failure rates for behavioral health by plan and bring the data to joint operating committees.
The 2027 rate backdrop
Compliance spending decisions do not happen in a vacuum. CMS finalized a 2.48% Medicare Advantage rate increase for 2027, which Healthcare Dive characterized as favorable to insurers. The CY 2027 Advance Notice had paired near-flat initial rates with modified risk adjustment methodology, according to regulatory analyses from McDermott+ and Sheppard Mullin, so the final figure read as relief rather than windfall.
That relief is thin against continued utilization pressure. Plans entering 2027 with slightly better rate math and persistent medical cost trend will be tempted to keep squeezing administrative lines. The Carelon matter is a reminder that network data operations is one line where cutting generates a liability that shows up later with interest.


