Every patient asking about a new medication has the same question. Dr. Robert Granko, System Vice President, Pharmacy Business Operations at UNC Health, puts it plainly. “Can I get my medication, and can I get it in time at a cost I can afford?”
That question lands on pharmacy teams every single day. At UNC Health, the team manages nearly 50,000 referrals a year, including 40,000 prior authorizations, with a 74% first-pass approval rate. This year, with AI embedded in the workflow, the team absorbed more than 45% year-over-year volume growth without proportional headcount increases.
Before deploying any technology, Granko asked the more important question. Not how do we keep up with volume, but what should this work actually look like?
That reframe is harder than it sounds. Most health systems respond to rising PA volume the way they respond to most operational pressure: add people, add hours, ask the team to do more. It works until it doesn’t. For many pharmacy teams, it already isn’t working.
“We didn’t deploy AI into existing processes unchanged,” Granko says. “We first and have continually asked, while challenging existing held beliefs, what should this work look like if the goal is timely, reliable patient access?”
The work at UNC has been about scale and visibility. UNC’s central pharmacy team sits at the core of how UNC coordinates medication access across prior authorization, financial assistance, specialty pharmacy, home infusion, ambulatory care, and 340B. These are not separate functions in Granko’s view. From a patient’s perspective, they are all part of the same journey.
When UNC expanded into infusion prior authorizations, what they found was revealing. Different submission methods, portals, fax, and EHR workflows had created a fragmented picture around the same patient. Duplicated work. Inconsistent tracking. Variable turnaround times. The fragmentation was not fully visible until they tried to standardize across it.
The data told a clear story. Even with strong performance and significant efficiency gains from AI, the current workflow design could not meet total system demand without change. That gap did not just surface a capacity problem. It brought leaders to the table. Pharmacy, providers, nursing, revenue cycle, IT, and strategy sat together in a two-day Express Workout to define what a future-state model should look like.
The visibility changed the conversation. As Granko puts it, “Our data didn’t just highlight a capacity issue. It’s helping us align leaders around a shared responsibility to patient access and re-imagine new possibilities.”
UNC is a large academic medical center forecasting 45%+ volume growth in its first year of this work. For all the scale it operates at, the conclusion Granko reached translates well beyond any single system. These solutions, he says, are labor complements rather than labor substitutes.
The value is not in doing the work faster. It is in what teams can do with the capacity they get back.
When the administrative burden lifts, capacity flows to patients. At UNC, that means faster outreach, proactive follow-up on stalled authorizations, and stronger financial navigation support.
The standard Granko holds is simple. “Efficiency gains only matter if patients and our providers feel them.”
Not review time in isolation. Not FTE counts. Whether the patient starts therapy this week or waits another two. Whether the provider gets a clear answer or another callback. Whether the pharmacy team ends the day having done the work they actually want to do.
UNC is still building. Granko is direct that expanding into infusions is early and there is more work ahead.
The question Granko has moved past is whether to redesign. That decision has been made. The real debate is no longer whether pharmacy operations need redesign. The more important question is how much better pharmacy can perform when workflow, technology, and operating models are aligned to support the team and patients.
UNC Health uses Latent Health’s enterprise pharmacy intelligence platform, which enabled their AI-powered prior authorization.
At the Becker's 11th Annual IT + Revenue Cycle Conference: The Future of AI & Digital Health, taking place September 14–17 in Chicago, healthcare executives and digital leaders from across the country will come together to explore how AI, interoperability, cybersecurity, and revenue cycle innovation are transforming care delivery, strengthening financial performance, and driving the next era of digital health. Apply for complimentary registration now.
