Hospital and insurer price files match exactly on 10.8% of rates, Trilliant finds
A new study of the price data hospitals and health plans must publish finds the two sources often list different prices for the same service at the same hospital, and that commercial rates for one surgery can vary more than eightfold.

Hospitals and health plans list the same negotiated rate in their public price files only 10.8% of the time, according to a study1 that health data company Trilliant Health published on November 10. Across every hospital, payer and procedure it compared, 17.9% of the paired rates were less than half or more than one and a half times each other.
Why it matters
Federal rules require hospitals to post the rates they negotiate with each insurer, and require most health plans to post the same rates in machine-readable files under the Transparency in Coverage rule. In principle, the two should describe the same contracts. When they disagree, an employer checking what its plan pays, or a patient trying to estimate a bill, cannot tell which figure is right without other data. The study's authors conclude that the conflicts limit how far the files can be compared or used.
The study also shows how wide the range of prices behind those files is. A procedure that costs a few thousand dollars at one site can cost several times as much at another, and the price a plan pays shows little relationship to how patients fare.
Where the files disagree
The gaps followed patterns by insurer and procedure. For three common endoscopy codes (CPT 43239, 45380 and 45385), the rates in Aetna's files averaged about 1.7 times what hospitals posted for the same services, the study found. For gallbladder removal and hernia repair, Aetna's figures were close to the hospitals' own. UnitedHealthcare's figures sat near or slightly above hospital-posted rates for the endoscopy codes and slightly below them for the laparoscopic surgeries.
The largest single-hospital gaps were large in dollars. For one colonoscopy code, an Aetna rate was $11,001 higher than the hospital's posted rate. For gallbladder removal, a hospital posted a UnitedHealthcare rate $31,048 higher than the plan's own file showed.
How far prices range
Within a single insurer, inpatient prices spread widely. UnitedHealthcare's rates for a coronary bypass case (MS-DRG 236) ran from $28,861 to $195,609, a 6.8-fold range. Aetna's rates for a hip or knee replacement (MS-DRG 470) ran from $13,290 to $117,081, or 8.8-fold.
Site of care mattered as well. UnitedHealthcare's median rate for arthroscopic rotator cuff repair (CPT 29827) was $13,477 at hospital outpatient departments and $5,492 at ambulatory surgery centers, according to the study.
Prices also moved unevenly over time. For one breast surgery code (CPT 19307), commercial rates rose faster than general inflation of 6.6% from 2024 to 2026 in 26 of 47 metro areas. The change ranged from a 24.5% drop in Oklahoma City to a 45.4% increase in San Francisco.
Across four common hospital admissions, including heart failure and pneumonia, the correlation between price and mortality ran from negative 0.30 to positive 0.25, the study found, which the authors describe as little relationship between what a hospital charges and its outcomes.
Context
The study draws on both kinds of files, mostly from 2026, with trend comparisons back to 2024. It names one problem in particular: "zombie rates," prices listed for providers who do not actually perform the service. It says federal amendments to the Transparency in Coverage rule made in October 2026 target those entries.
The authors consider plan files generally more reliable because they update more often and cover more types of providers. They note that neither source reports how many times a service was performed, so the files are often combined with claims data. The study does not say how many hospitals it included in each comparison.
What comes next
No regulator has responded to the study. The question to watch is whether federal enforcement of the hospital and plan rules begins to test the files against each other, and whether the October amendments reduce the gaps when plans publish under them.



