The Directory Is Lying to You
What ghost networks actually cost Hawaiʻi patients and why no one is held accountable.
by David Isei, MPH, MAOL, PMP, Executive Director, Hawaiʻi Healthcare Task Force, May 24, 2026
You have insurance. You open your plan's provider directory and see dozens of physicians near your zip code. You call the first one. Disconnected. You call the second. No longer accepting your plan. The third hasn't been accepting new patients for 2 years. By the time you reach someone who can actually see you, weeks have passed, and the condition you needed treated has progressed.
It is not a bad luck story. That is the design of the system working exactly as built.
What the Data Shows
The Hawaiʻi Health Intelligence Platform (HHIP) analyzed provider directory listings across a March 2026 cohort of 412 sampled entries. Forty-two percent failed basic reachability validation. The phone didn't connect, the provider no longer worked at the listed address, or the practice had no record of the plan affiliation shown in the directory. Just 31% of sampled providers were verifiably accepting new patients. That means 69% of the listed providers were functionally unavailable to someone holding an insurance card issued by those same plans.
It is not a rounding error. This is the majority.
The Federal Picture Confirms It
In October 2025, the U.S. Department of Health and Human Services Office of Inspector General published findings on behavioral health provider networks in Medicare Advantage and Medicaid managed care plans (OIG, October 2025). On average, 55% of behavioral health providers listed in Medicaid managed care plan directories did not provide any care to enrollees. In some plans, that figure exceeded 60%. At least one-third of listed providers were inactive across the majority of Medicare Advantage plans reviewed.
The OIG also found that the average Medicare Advantage plan contracts with only 16% of behavioral health providers in its service area. The directory, then, is not merely inaccurate. The underlying network is inadequate to begin with, and the directory overstates it.
Hawaiʻi's Own Review Found Total Failure
Hawaii's 2024 External Quality Review, conducted by HSAG for the Med-QUEST Division, assessed all five QUEST Integration health plans against the Access and Availability network adequacy standard. All five plans failed to meet the standard for any indicator (HSAG/MQD, 2024 EQR).
Not one plan. Not most plans. All five. Every plan that covers the majority of Hawaiʻi's Medicaid population failed every access and availability metric. That result did not generate emergency regulatory action. The plans continued operating. Patients continued enrolling. The directories continued listing providers who were not there.
The Physician Count Makes It Worse
Even a perfectly accurate directory would tell a grim story. Of the 12,688 licensed physicians in Hawaiʻi, only 3,647 are currently providing patient care, according to a January 2026 report by Honolulu Civil Beat. Fewer than one in three licensed physicians is actively seeing patients. When part-time practice is accounted for, the full-time equivalent count in 2025 was just over 3,000. That is 644 physicians short of documented demand. Factored for geography, the shortfall reaches 833.
Maui County is short 45 primary care physicians. The Big Island needs 224 additional physicians. These are not projections. These are current, documented gaps.
A directory listing 12,000 physicians means nothing when fewer than 4,000 of them are actually treating patients.
This Is Not a Data Quality Problem
The framing that gets applied to ghost networks is almost always technical. Plans need better data hygiene. Directories need more frequent updates. Verification workflows need improvement. Federal law already requires plans to validate directory data at least every 90 days under the No Surprises Act. Yet research published in the American Journal of Managed Care found that 40% of identified inaccuracies persisted for an average of 540 days, and only 13% of previously flagged inaccurate listings were ever fully corrected (AJMC, 2025).
The law exists. The requirement exists. The inaccuracies persist anyway. That is not a data quality problem. That is a compliance theater problem.
The Accountability Gap Is Structural
Here is how network adequacy is actually measured in most states, including Hawaiʻi: plans submit directory data to regulators. Regulators count the listings. If the count meets the threshold, the plan passes. No one calls the providers. No one verifies that the listed address is still in operation. No one checks whether the provider has any patients enrolled in the plan.
The standard is satisfied by the existence of a name in a spreadsheet.
HHIP's reachability validation work is, to my knowledge, the only independent infrastructure in Hawaiʻi doing something different. We contact the providers. We verify the listings against actual availability. We test the claims. What we consistently find is that the gap between what plans report and what patients actually encounter is not a margin of error. It is the rule.
Plans have no systemic incentive to shrink their directories. Smaller directories invite regulatory scrutiny and potentially trigger network adequacy failures. Inflated directories pass compliance reviews. The rational move, under current rules, is to list more providers, verify fewer of them, and let the patient find out the hard way.
What Needs to Change
Network adequacy must be measured by verified encounters, not directory submissions. Regulators need independent validation capacity, not plan-submitted counts. Hawaiʻi's Med-QUEST Division and the Insurance Division should require plans to demonstrate, with claims data, that listed providers actually served enrollees in the prior measurement period. Providers who generated zero encounters for a plan's enrollees over a defined window should be removed from that plan's directory.
That is a solvable technical problem. It requires political will, not innovation.
The patients who are failing to get appointments are not trying hard enough. They are encountering a system that counts phantom access as real access, submits that count to regulators, and then collects premiums for coverage it cannot deliver.
The directory is not a guide to care. Right now, in too many cases, it is a document designed to pass a test. Patients deserve to know that. And regulators should be required to prove otherwise.
---30---
David Isei is the Executive Director of the Hawaiʻi Healthcare Task Force and Architect of the Hawaiʻi Health Intelligence Platform (HHIP). HHIP is the only independent intelligence infrastructure analyzing Hawaiʻi's healthcare system.