Rural Colorado Hospital Uses Private AI to Recover Unpaid Claims

Spanish Peaks Regional Health Center, a nonprofit facility in rural Huerfano County, Colorado, faces a problem common to healthcare providers nationwide: insurance companies deny or delay payment on roughly 10% of submitted medical claims. For a facility operating a 20-bed hospital, 90-bed nursing home, and several clinics, those unpaid claims can total hundreds of thousands of dollars annually.

Chief Financial Officer Daron Hashir, who joined Spanish Peaks in February, has launched a trial with Denver-based Iterate AI to address the revenue gap. The system uses agentic artificial intelligence—software that not only analyzes data but takes action—to identify billing errors, missing payments, and what Hashir calls "ghost denials," where insurers withhold payment without issuing a standard denial code.

How the private AI system works

Iterate AI's platform operates entirely within corporate firewalls, keeping sensitive medical data on-premises rather than sending it to external servers. The system creates specialized software agents that analyze denial documents, match claims against insurer responses, and draft appeals letters complete with contractual language—all in minutes.

"They're having to go into the claims, which may be 20 pages long. And they're trying to find why the payers are denying it," said Kevin Homer, Iterate's vice president of sales. "Insurance companies have stopped using the words denied or denial because they're too easy of keywords."

The technology addresses a volume problem that human staff cannot match. One large southeastern healthcare provider sent Iterate 10 million rows of claims data representing $27 billion in billings—far more than a typical revenue cycle department of 20 to 25 people can manually review.

Why it matters

The stakes extend beyond hospital balance sheets. Industry data shows initial claim denials by insurers rose from 10.2% in 2021 to 11.6% in 2024, contributing to $48.4 billion in lost revenue for billable services last year—a 25% increase from the prior year, according to health-tech consultancy Kodiak. For rural facilities like Spanish Peaks, where patients who cannot afford care still receive treatment, recovering denied claims "prevents us from having to write it all off," Hashir said. Individual denied claims at the facility average between $5,000 and $25,000.

Privacy protections in healthcare AI

Private AI systems like Iterate's address data security concerns by operating within existing corporate infrastructure rather than routing information through external cloud services. Alex Kimata, a data privacy attorney at Holland & Hart's Boulder office, said the approach requires the same fundamental protections—encryption, access controls, vendor vetting—that companies should apply to any sensitive data system.

"This idea of I just want to be an AI company quote-unquote, doesn't really make sense," Kimata said. "This should be intentional and it should be customized to how you want to use AI, including what vendor you use, what program you use and what controls you put on it."

Iterate AI, founded in 2013 by eBags cofounder Jon Nordmark, originally built revenue optimization tools for retailers like Ulta Beauty and Circle K. The company's healthcare pivot leverages recent advances in agentic AI that can execute tasks autonomously rather than simply generating text responses.

While Iterate's trial at Spanish Peaks analyzed only partial historical data, Hashir said the results demonstrated the technology's ability to identify issues quickly and accelerate the appeals process. She is now working with Iterate to develop an additional AI agent that will identify physicians who need training on accurate coding practices.

Details of the Spanish Peaks trial were first reported by The Colorado Sun.