Utah tests AI to renew prescriptions without doctors

Utah has started a pilot scheme that allows an artificial intelligence system, rather than a clinician, to renew certain prescriptions for patients with chronic conditions, in what industry observers describe as a first for the US.

The programme, developed with health technology company Doctronic, automates routine repeat prescriptions for a restricted list of medicines. State officials view it as a response to rising healthcare costs and workforce shortages, while critics highlight safety and regulatory risks.

Under the initiative, the AI system handles renewals without direct involvement from a doctor once its performance has been validated in each medication class. The move marks an expansion of AI's role in clinical decision-making beyond diagnostic support and administrative triage.

How it works

Patients using the service access a webpage that verifies they are physically in Utah. The system then pulls the patient's prescription history and presents a list of medications that qualify for renewal under the pilot.

The AI then guides the patient through a series of structured clinical questions. The questions mirror those a physician would use when assessing whether a refill remains appropriate for a chronic condition.

If the system clears the renewal, it sends the prescription directly to a pharmacy. No human clinician reviews those renewals once the AI has passed an initial oversight phase for that medication category.

The pilot covers 190 commonly used medicines. It excludes pain management drugs, ADHD medications and injectables, which officials classify as higher risk.

Doctronic will charge patients USD $4 per prescription renewal during the early phase of the programme. Co-founder and co-CEO Matt Pavelle said the firm expects the cost structure to change as volumes rise and as insurers or subscription models take on more of the expense. "It's hard to get a renewal - if you have a chronic condition and you can't get your medication, terrible things happen," said Pavelle.

State's rationale

Utah officials present the scheme as a targeted response to pressure on health systems, especially in underserved areas. They argue that automating repetitive clinical workflows can free physicians for more complex tasks and cut delays in access to long-term medication.

Margaret Busse, Executive Director of the Utah Department of Commerce, said healthcare costs continue to escalate and clinicians in rural regions face growing demands. The state sees AI-based renewals as a way to ease that strain and lower patient costs.

"It's also a way to "provide a pathway to innovation for entrepreneurs who are using AI in creative ways that may be bumping up against regulation," she said.

Busse said the state expects the company to build confidence in the system among patients. "The company has to do that kind of trust building with their patients," said Busse. "We want it to be done in such a way that people will trust that Utah is looking at this carefully and is not being cavalier about how we granted this regulatory mitigation. In a way it's a risk for us as we do this."

Medical concerns

Doctors' groups and some regulators have raised concerns about the wider use of AI in prescribing. They highlight risks around missed diagnoses, subtle signs of deterioration and complex drug interactions.

Dr John Whyte, CEO and Executive Vice President at the American Medical Association, warned about the implications of removing clinicians from parts of the prescribing process. "While AI has limitless opportunity to transform medicine for the better, without physician input it also poses serious risks to patients and physicians alike," said Whyte.

Clinicians and regulators also point to the risk of misuse. Automated systems could be targeted by people seeking inappropriate access to controlled medicines or by those attempting to exploit gaps in the screening logic.

Al Carter, CEO and Executive Director at the National Association of Boards of Pharmacy, said pharmacists already use AI in prescription fulfilment and consultations. "The one challenge from a board of pharmacy standpoint is how do you regulate all this technology, and is this technology good for health care?" said Carter.

System performance

Doctronic has presented Utah regulators with internal data that compares its AI system with human clinicians across 500 urgent care cases. The company said the AI's treatment plan matched physicians' decisions in 99.2 per cent of those cases.

Co-founder Dr Adam Oskowitz, an Associate Professor of Surgery at the University of California San Francisco, said the system runs medication and safety checks that often exceed routine human workflows. "The AI is actually better than doctors at doing this," said Oskowitz. "When you go see a doctor, it's not going to do all the checks that the AI is doing."

Oskowitz said the model is designed to err on the side of caution. The system escalates any case that triggers uncertainty flags to a human physician. Doctors will review the first 250 prescriptions in each medication class that the AI issues. Once those pass review, subsequent renewals in that class proceed autonomously.

Doctronic has also arranged a malpractice insurance policy that explicitly covers its AI system. The company says the arrangement aligns the AI's legal responsibility with that of a doctor who prescribes medication.

"In medicine, there's always going to be potential issues that patients have," said Oskowitz. "Whether it's caused by the AI or not - we will take the risk. I think this is going to be infinitely safer than a human doctor."

Regulatory grey area

The Utah initiative highlights uncertainty over which regulators have primary authority over AI systems that perform clinical tasks. States have long overseen the practice of medicine, while the Food and Drug Administration regulates medical devices and certain software.

Lowell Schiller, a former FDA chief counsel, said the Doctronic system could fall within state jurisdiction because it renews prescriptions and therefore functions as a practitioner. At the same time, he noted that the FDA has stated it can treat AI as a medical device when it is used to diagnose, treat or prevent disease.

Schiller said the agency could hold off on intervening, following a similar pattern to its approach to medical marijuana where it has deferred to states. If the FDA later determines that the system is being marketed without the necessary authorisation, it could seek changes to bring the product into compliance.

The agency recently signalled a stricter stance on health-related technology. It wrote to wearable maker Whoop and said the company could not market blood-pressure estimation features without FDA clearance.

Zach Boyd, Director of Utah's Artificial Intelligence Policy Office, said the traditional split between state and federal roles is becoming more complex. "Now we're in this weird place where there are devices - maybe you could call them devices - that are purporting to practice medicine," said Boyd. "Our philosophy has been to just take care of our side, of the state's authority, and the FDA is going to figure out what it's going to figure out."

Doctronic's founders are in discussions with other states including Texas, Arizona and Missouri about extending the model. They are also exploring a national regulatory pathway that would avoid negotiating separate rules in each jurisdiction.