For rural healthcare providers, retaining clinical talent is a top challenge. Physicians are often the most expensive employees to recruit and the primary revenue generators. Losing a provider to burnout means absorbing immense sunk costs and potentially losing patients to competitors.
“There’s an immense amount of sunk cost when you have that turnover, just what you’ve invested directly into the physician, but then also you could potentially lose those patients in the market that then impact the ecosystem of the healthcare system,” explains Michael Lewis, CFO of Onvida Health, a health system located in rural southern Arizona.
To prevent this, Onvida Health leadership targeted “pajama time,” the burdensome, after-hours documentation work that frequently drives physician burnout. The organization partnered with Ambience Healthcare to launch an AI-driven documentation tool across multiple disciplines, including allergy and immunology, anesthesia, cardiology, endocrinology, family medicine, and GI.
Reducing Burnout and Reclaiming Capacity
The platform generated an immediate 30% reduction in after-hours documentation, leading one clinician to joke that they needed a new hobby to occupy the extra free time, according to Lewis. When a patient left the exam room, “the majority of the documentation is done, and then at the end of the day, the physician is done,” Lewis says.
The health system has calculated a positive impact of about $24,000 per physician since it began using ambient AI. Some of the ROI comes from the reduced need for rework while the majority comes from improved patient access.
Because the technology automatically drafts comprehensive notes by listening to the patient encounter, highly productive physicians have gained significant daily capacity. “On average, one additional visit per day,” Lewis says, noting that this increased volume occurred without expanding operational hours, adding new physical space, or hiring additional staff.
The platform also captures clinical complexity that time-crunched physicians might not document. For instance, instead of a brief, bulleted note stating an MRI was reviewed, the AI drafts a complete paragraph detailing specific findings, such as a torn meniscus, and the subsequent care plan.
“It’s made the documentation match the care that’s being rendered so that we can appropriately and compliantly capture the higher code,” Lewis notes.
This streamlined documentation process has directly accelerated Onvida’s cash flow, generating a 20% to 25% improvement in charge lag. Since documentation is completed more quickly, revenue cycle teams can bill more quickly.
While the implementation is still in its first year, Lewis anticipates that future data will show corresponding drops in downstream denials and authorization friction because initial clinical documentation is far more complete for payers to review.
Navigating Adoption Challenges
Despite the clear benefits, implementing the platform required overcoming skepticism among clinicians. During the initial rollout meeting with nearly 100 physicians, Lewis recalls that getting volunteers for the pilot program was “like pulling teeth,” resulting in leadership having to “volun-tell” specific doctors to participate.
The attitude around adoption quickly shifted. The pilot group became “believers” and “zealots” who shared their positive experiences with other clinicians until virtually all were eager to use it. “They went from doubting to now they can’t get it fast enough,” Lewis says.
Now, clinicians are using the technology for 84% of patient encounters, and the early success has led the health system to begin pilot testing the technology for inpatient hospitalists and ED doctors.
Choosing an Ambient AI Vendor
For a smaller rural health system, the margin for error on technology investments is incredibly tight. A large, highly-matrixed organization can absorb a million-dollar mistake much more easily than a critical access hospital.
To mitigate this risk, Onvida established specific metrics to evaluate an ambient AI tool’s success during a 60- to 90-day pilot. They also selected a vendor based on a willingness to actively partner and modify the technology to fit specific clinical workflows, rather than forcing the adoption of an off-the-shelf application. When early coding issues were identified, the vendor provided an “all hands on deck” approach to fix the platform, Lewis says.
By demanding rigorous early ROI metrics and insisting on active partnership with its vendor, Onvida successfully navigated the transition, proving that rural health systems can embrace adoption of emerging technologies.
For revenue cycle leaders, Onvida’s success shows how clinical workflow automation can drive measurable ROI, improve physician satisfaction, and increase patient access.