15 to 20% of all first-time claims are denied, 65% of those denials are never resubmitted and 86% were entirely avoidable. The math adds up to $36 billion in annual revenue losses for U.S. health systems, plus an additional $20 billion spent trying to recover it.
Most of that loss can be traced back to the exam room. For revenue cycle leaders, that makes clinical documentation not just a physician problem but a system wide one
During a featured session at Becker’s 16th Annual Meeting in April, sponsored by Suki, Umar Bowers, MD, internal medicine physician and owner-operator of three clinics in Wilmington, North Carolina and Katie Prevas, PA-C medical director at Suki, physician assistant and former director for growth and innovation at MedStar Health Urgent Care in Washington D.C., discussed how ambient clinical intelligence is changing what happens at the point of care and what that means for coding accuracy, provider retention and downstream revenue.
Below are four takeaways from their conversation.
1. The last-mile problem
The session opened with a pertinent analogy from the retail sector where Amazon spent decades perfecting methodology of procurement and shipping from factories and warehouses Yet 53% of its total logistics cost comes from the van, the driver and the front door.
Healthcare RCM has the same dynamic. billing systems, clearing houses and denial management tools have all improved claim adjudication but the exam room is where most revenue challenges originate. Clinicians under time pressure,and fearful of audit, document at lower E&M levels than the visit supports. They also miss codable secondary diagnoses and lose disclosures that happen at the door.
“Charting three days later, you’re going to do the minimal amount of work just to close the note,” Dr. Bowers said, “It doesn’t accurately reflect all the things we do.”
2. Changing workflows
Both panelists described the overwhelming loss of information when ambient wasn’t in use. For Ms. Prevas, secondary diagnoses — a patient’s smoking history, family history of heart disease, relevant comorbidities — were clinically relevant but rarely made it into the chart. “All of this is codable and pertinent to the robustness of the documentation,” Ms. Prevas said, “I would just write in my diagnosis and leave it at that.” For Dr. Bowers, the shift was in HCC coding accuracy, critical in value-based care arrangements where risk adjustment factor scores determine how complex, and therefore how costly, a patient population is expected to be. Critically, neither panelist changed how they spoke with patients to achieve these gains. The technology meets clinicians where they are.
Ms. Prevas came to Suki as a user before joining the company. While serving as director of innovation for a health system operating 33 urgent care clinics across the D.C. tri-state area, she was tasked with finding a documentation solution that didn’t rely on live scribes, a model that wasn’t economical or sustainable at that scale. After implementing Suki across the network and seeing the results firsthand, she was drawn to the company’s clinician-centricity.
3. Painting the right picture
A recurring theme was the distinction between coding accurately and upcoding which ambient intelligence makes easier to navigate. Dr. Bowers emphasized that in value-based care, undercoding is just as damaging as overcoding: if a complex patient’s chart doesn’t reflect their actual acuity, shared savings calculations break down and the health system absorbs costs it never anticipated.
Ambient tools with bidirectional EHR integration can read the full problem list in real time, provide in-workflow nudges and give clinicians positive reinforcement when documentation supports the coded level.
“I don’t feel like I’m trying to play a guessing game,” Ms. Prevas said. “I can feel secure in my documentation.”
4. The true ROI
The panelists made a case that the ROI of ambient intelligence extends well beyond coding. Dr. Bowers described using the technology to reduce his own end-of-day note burden — enabling him to be present at his children’s sports practices rather than mentally cataloging 30 unfinished charts.
Ms. Prevas’ experience found that documentation time dropped from roughly five minutes per note to one to two minutes, saving clinicians an average of 90 minutes on a 12-hour shift and eliminating the off-day charting that had become a retention risk.
Dr. Bowers noted that one of his older providers — a 63-year-old PA who had considered leaving — extended her career because the tool reduced her daily burden. “She bounces out of the office right at 5 o’clock and has joy again,” Dr. Bowers shared.
The panelists closed by looking ahead. The next wave of ambient intelligence, both argued, moves beyond note generation into order entry — verbal commands that generate labs, imaging requests and referral letters without additional keystrokes.
For health systems still evaluating the technology, the case is increasingly straightforward. The clinical documentation gap is also a revenue gap, and the tools to close it are already available.
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.