Ghost Networks Persist in Commercial Insurance Mental Health Plans

Fewer than one in four calls that reached a definite outcome ended with an appointment with the psychiatrist or psychologist listed in a commercial insurer's online directory, according to a secret-shopper study published Oct. 16 in JAMA Health Forum. Researchers at Ohio State University and Stanford University placed 5,822 calls between October 2025 and March 2026 to clinicians listed by the largest commercial insurer in Delaware, Montana, New Jersey, Rhode Island and Virginia.

Among the 2,252 calls that reached a clear outcome, 23.6 percent produced an appointment with the listed clinician. Counted against every first attempt, the figure fell to 12.4 percent for an appointment with any clinician. Directory errors — wrong contact information, wrong specialty, a practice at capacity — affected 55.6 percent of the evaluable calls.

The authors, Simon Haeder, Yuqin Lou, Wendy Xu and Benjamin F. Miller, concluded that access barriers previously documented in public insurance programs also extend to privately insured patients, and that "commercial coverage did not translate into ready access to mental health care."

Six in ten calls never reached a person

The largest single finding was not a failure but a silence. Of 5,822 calls, 3,570 — 61.3 percent — could not be evaluated at all, mostly because the caller reached an answering machine or an automated system with no option to speak with staff. Any call that left no way to schedule was excluded from the appointment and error rates, because the cause could not be determined.

The authors noted that many mental health clinicians work in solo or small-group practices without front-desk staff, which limits their capacity to answer a telephone during clinical hours. Callers made a single attempt per listing and did not leave messages, so a persistent patient working through a call-back list would likely do better than the audit's headline rate suggests.

Wrong numbers more often than full practices

Where a person did answer, the problem was usually the listing rather than the clinician's availability. Wrong contact information accounted for 31.0 percent of evaluable calls, including 257 calls that reached disconnected or out-of-service numbers. Wrong specialty listings accounted for 22.2 percent. Capacity limits — a clinician simply not accepting new patients — blocked scheduling in another 21.6 percent.

When appointments with a different clinician at the same office number were counted, the success rate rose to 32.1 percent. Among callers who secured a specific date, the average wait was 20.5 days, with psychiatrists slower than psychologists at 22.5 days against 17.3.

Performance varied by plan, but no carrier performed well. Horizon Blue Cross Blue Shield of New Jersey posted the highest appointment rate among evaluable calls at 32.9 percent. CareFirst BlueCross BlueShield in Virginia posted the lowest at 18.0 percent and the highest directory error rate at 70.8 percent, though that estimate rests on only 42 evaluable calls.

The drive is part of the wait

Distance compounded the problem. Average travel time to an appointment was 17.1 minutes overall, but 41.8 minutes in Montana and 30.8 minutes for rural residents across the five states. For weekly therapy, that drive sits alongside the wait rather than instead of it.

All of the carriers surveyed were Blue Cross Blue Shield licensees, chosen because each held the largest commercial market share in its state. The authors argued that if the largest carriers cannot keep directories accurate, smaller carriers are likely to fare worse, while conceding the study did not test that proposition.

A rule that already exists

The federal No Surprises Act requires insurers to verify and update directory information at least every 90 days. The authors wrote that errors persisting more than three years after those provisions took effect suggest the current rules are either insufficient or not meaningfully enforced.

The finding is not new to Washington. A 2023 Senate Finance Committee secret-shopper study of Medicare Advantage plans found staff could make an appointment only 18 percent of the time, and then-chair Ron Wyden called the "widespread existence of ghost networks" unacceptable. The JAMA Health Forum authors also cite a 2009 study in which 45 percent of calls to mental health clinics reached only an answering machine — a share close to the 44 percent observed in this audit, which suggests the problem has survived well over a decade of attempted fixes.

The study does not show that any insurer broke the law. It measures the distance between the clinicians a directory lists and the clinicians a patient can actually book.

What the audit does not establish

The design is a snapshot rather than a measure of outcomes. It covered five states, one standard PPO plan per carrier, and a single call attempt per listing. It cannot say why so many offices were unreachable. Results may not transfer to other insurers, plan types or states.

Funding came from Inseparable, a mental health advocacy nonprofit that took part in discussions about which states to study but had no role in conducting the study, analyzing the data or preparing the paper, according to the authors. One author disclosed personal fees from law firms for work outside the study.

For patients, the study's practical advice runs against the directory habit. Calling several listed clinicians rather than one at a time, leaving messages, and asking directly whether a clinician is in-network for a specific plan and accepting new patients will surface answers a website does not. Keeping a written log of names, dates and outcomes creates a record that a plan can be asked to act on. Under the No Surprises Act, a patient who relied on inaccurate directory information about network status is generally entitled to in-network cost sharing.

Anyone in crisis should not wait on a directory search. The 988 Suicide and Crisis Lifeline answers calls and texts around the clock, and telehealth services can be a faster entry point for people who cannot find an in-person opening. Outpatient programs and medication-assisted treatment providers are worth calling directly, because a clinic's own intake line is a different channel from an insurer's directory.