AI Prior Authorization and Denials: A Surgeon's View

Most conversations about AI prior authorization start in the back office, with automated submissions, faster status checks, and smarter queues. Those matter. But they all depend on something that happens much earlier, in the exam room, the moment a physician documents the visit.

If the clinical details aren’t in the note, no amount of downstream automation can put them there.

That’s the core argument of EP 09 of Truly Integrated. Dr. Matthew Tomaino, MD, MBA, FAAOS, Medical Director of Orthopedics and Sports Medicine at WellNow Urgent Care, joined iScribe CEO Pat Williams to explore AI’s role beyond documentation, in prior authorization, denials, and the revenue cycle. He brings more than three decades as a shoulder, hand, and elbow surgeon, including years running his own solo practice, Tomaino Orthopaedic Care in Rochester, New York.

Prefer to listen? Hear Dr. Tomaino’s full conversation on Truly Integrated, EP 09: episode page · Apple Podcasts · Spotify

Key Takeaways

• AI-generated documentation must be concise, clinically relevant, and reflective of the physician’s own judgment.
• The note supports reimbursement, prepares the physician for future visits, and protects the integrity of the medical record.
• Real-time AI guidance that flags missing clinical details is a powerful lever for preventing prior authorization problems and claim denials.
• Ambient AI reduces after-hours documentation and returns the physician’s full attention to the patient.
• Automating coding and operative notes accelerates accurate claims and reduces administrative cost.
• The ROI of AI includes job satisfaction, reduced burnout, and regained personal time on top of revenue.

Why Prior Authorization Is a Documentation Problem

Prior authorization is supposed to confirm that a service is medically necessary. In practice, approval hinges on whether the documentation proves it through the history, the exam findings, the failed conservative treatments, the imaging, and the reasoning that connects them.

When any of those elements is missing or vague, the result is a delayed authorization, a request for more information, or a denial. Staff chase the physician. The physician reopens a chart from last week. The patient waits.

The volume makes every gap expensive. In an AMA survey of 1,000 physicians in late 2024, practices reported completing 39 prior authorization requests per physician, per week, and physicians and their staff spent an average of 13 hours each week completing them.

This is why our mission at iScribe is to end documentation-related denials, which we see as a fundamental flaw of modern medicine. The cause is a system that asks physicians to document perfectly against constantly changing rules, in real time, while seeing a full clinic.

Medicare Advantage data shows how often the first answer doesn’t hold. In 2024, Medicare Advantage insurers fully or partially denied 4.1 million prior authorization requests. Only 11.5% of those denials were appealed, but 80.7% of the appeals were partially or fully overturned, according to KFF. KFF notes this could mean the initial request should have been approved, or that it was missing the documentation needed to justify the service. Either way, patients may have waited for care that was ultimately approved.

The Note Has to Sound Like the Physician

Dr. Tomaino has authored more than 150 scientific articles and book chapters, served as Chief of the Division of Hand, Shoulder, and Elbow Surgery at the University of Rochester Medical Center, and earned an MBA from the University of Pittsburgh. He understands both the clinical record and the business that depends on it.

His standard for AI documentation is clear. Notes must be concise, clinically relevant, and an accurate reflection of the physician’s judgment. A bloated note full of transcribed small talk is just as dangerous as a thin one. It buries the medical necessity a reviewer needs to see.

That’s why iScribe produces structured, claim-ready notes using provider-specific macros, specialty-tuned to each physician’s preferences and requirements, with a processed note ready in roughly 30 seconds.

Where AI Prior Authorization Delivers Real Value

Dr. Tomaino identified where AI creates leverage across the revenue cycle, and his points add up to a practical roadmap for AI prior authorization and denial prevention.

1. Capture the encounter completely, at the source. Ambient AI documents the visit as it happens, so the clinical details a prior authorization depends on are recorded while they’re fresh, not reconstructed after clinic.

2. Flag what’s missing before it becomes a problem. Real-time AI guidance that identifies missing clinical details before submission is the difference between an approval and a denial. It’s also where iScribe is building. Real-time denial prevention capabilities are in development now.

3. Code accurately at note completion. iScribe Code Intelligence gives completed notes an E&M code recommendation with 95% audited accuracy, supported by detailed analysis of the AMA MDM guidelines. It’s available for orthopedic and urology practices on athenahealth or NextGen.

4. Automate the repetitive documentation. Dr. Tomaino highlighted the value of automating coding and operative notes so practices submit accurate claims faster and reduce administrative expense.

5. Move the work out of the physician’s evening. Ambient AI reduces after-hours documentation. iScribe providers save roughly two to four hours per day.

How AI Prior Authorization Works: Back Office vs. Exam Room

Many AI prior authorization tools work after the visit, on the request. Point-of-care documentation works during the visit, on the evidence. They solve different problems.

• When it works → Back-office prior authorization tools: After the visit, once a request is built → Point-of-care AI documentation: During the visit, as the note is created.
• What it does → Back-office prior authorization tools: Checks payer requirements, submits requests, and tracks status → Point-of-care AI documentation: Captures the history, exam, prior treatment, and reasoning.
• What it depends on → Back-office prior authorization tools: What the note already contains → Point-of-care AI documentation: The physician’s conversation and judgment.
• What it can’t fix → Back-office prior authorization tools: Clinical details that were never documented → Point-of-care AI documentation: Payer rules and response times.

In our view, the two work best together, with the note doing the heavy lifting. A submission tool can only send what the documentation supports.

“It Refocuses Everything on the Room”

For all the talk of revenue cycles and authorizations, the point that stuck with us most was about attention.

“When you actually think about the end organ, you know, just like our hands experience the environment, the doctor experiences the patient,” he said. “But all this stuff out there that’s growing and growing and growing, tons of dollars putting into it, is not directly patient care… But what it does do is it refocuses everything on the room because your brain share is not now split up among all these other things. It’s focused just on the patient. The technology is doing the rest.”

That’s the promise of AI prior authorization done right. Faster approvals matter, and so does a physician whose full attention stays where it belongs.

The ROI Goes Beyond Revenue

Dr. Tomaino made a point that most ROI calculators miss. The value of AI isn’t limited to measurable revenue. Improved job satisfaction, reduced burnout, and regained personal time deliver significant returns, too.

For a solo or partner-owned practice, those returns compound. A physician who isn’t charting at 10 p.m. stays in practice longer, sees patients with more energy, and makes better decisions. Combine that with fewer denials and cleaner claims, and the case for surgical-grade documentation becomes hard to argue with.

Better Documentation Protects the Record, and the Practice

Dr. Tomaino also emphasized that strong documentation helps physicians prepare for future visits and protects the integrity of the medical record. The same note that supports a prior authorization today supports continuity of care next month and stands up to an audit next year.

We’d argue the note is the foundation of every AI strategy in the revenue cycle. If you’re seeing prior authorization play out differently in your practice, we’d like to hear about it.

What the New CMS Prior Authorization Rules Change, and What They Don’t

Federal rules are starting to speed up the payer side. Under the CMS Interoperability and Prior Authorization Final Rule (CMS-0057-F), impacted payers, including Medicare Advantage organizations and Medicaid and CHIP programs, face new process requirements generally beginning January 1, 2026. Impacted payers other than QHP issuers on the federal exchanges must send decisions within 72 hours for expedited requests and seven calendar days for standard requests, and impacted payers must give a specific reason for every denial. The rule does not apply to drugs.

None of this changes what a reviewer needs to see. A faster decision on a thin note is still a denial, now with a reason attached. A specific reason helps a practice resubmit, but the strongest request is the one that never needs resubmitting.

Frequently Asked Questions

How does AI help with prior authorization?

AI supports prior authorization by capturing complete clinical documentation at the point of care, flagging missing details before submission, and automating repetitive administrative steps. Approval ultimately depends on what the note contains.

Can AI reduce claim denials?

Accurate, complete, specialty-specific documentation is the foundation of a clean claim. iScribe AI produces claim-ready notes, and iScribe Code Intelligence delivers E&M code recommendations with 95% audited accuracy. Real-time denial prevention capabilities are in development.

What should AI-generated clinical notes include?

They should be concise, clinically relevant, and reflect the physician’s judgment, capturing the history, exam, decision-making, and plan that support medical necessity.

What is AI prior authorization?

It is the use of artificial intelligence to help prepare, submit, and track prior authorization requests. Some tools work on submissions and status checks, while point-of-care documentation AI captures the clinical evidence those requests depend on.

What documentation do payers usually need for prior authorization?

Requirements vary by payer and service, but reviewers generally look for the history, exam findings, failed conservative treatments, relevant imaging, and the reasoning that connects them. Check each payer’s policy for its specific criteria.

What are the new CMS prior authorization deadlines?

Under CMS-0057-F, impacted payers other than QHP issuers on the federal exchanges must send decisions within 72 hours for expedited requests and seven calendar days for standard requests, generally beginning January 1, 2026. The rule does not apply to drugs.

Why are so many prior authorization denials overturned on appeal?

In 2024, 80.7% of appealed Medicare Advantage prior authorization denials were partially or fully overturned, according to KFF. KFF notes this may mean some requests should have been approved initially, or that they were missing the documentation needed to justify the service.

About Dr. Matthew Tomaino

Matthew M. Tomaino, MD, MBA, FAAOS, is a board-certified orthopedic surgeon and Medical Director of Orthopedics and Sports Medicine at WellNow Urgent Care, where he oversees musculoskeletal care at locations in Illinois, Indiana, Michigan, New York, and Pennsylvania. He founded his private practice in Rochester, New York, in 2008, after serving as Professor of Orthopedics at the University of Rochester Medical Center and, before that, on the orthopaedic surgery faculty at the University of Pittsburgh Medical Center, where he was Chief of Microsurgery. He earned his medical degree from the University of Rochester School of Medicine, trained in orthopedics and hand surgery at the University of Rochester, and completed a reconstructive microsurgery fellowship at the University of Toronto.

Hear the Full Conversation

Listen to Dr. Matthew Tomaino on Truly Integrated, EP 09, “Beyond Documentation: AI’s Role in Prior Auth and Denials.”

Defend your bottom line from the first word of the note. Start a complimentary 14-day trial of iScribe AI, fully set up before go-live with zero schedule reduction.