Providence has now expanded on previous studies by performing a comprehensive evaluation of the associations between an ambient AI system (Dragon Ambient eXperience [DAX]; Nuance) and clinician productivity and efficiency. The research team assessed the association between ambient AI use and objective documentation burden, after-hour documentation, and work volume among clinicians using retrospective EHR encounter metadata between July 1, 2023, and March 31, 2025. By the end of their study period, approximately 8% of clinicians within the health system were considered active users.

In a conversation with Healthcare Innovation, the research team discussed their findings. 

In their paper, the researchers found productivity outcomes demonstrated statistically significant differences between the pre-active and post-active ambient AI use periods. For instance, they found a significant decline in mean time spent on notes during the first month of ambient AI use. 

The Clinical Efficiency Profile (CEP) is an efficiency metric that is generated by Epic itself. The researchers found that the clinicians’ mean CEP scores had no immediate or sustained association with ambient AI use. “I think that it was important with the study to look at efficiency and productivity and administrative burden from a variety of different lenses, because there's not one way or even one standard on how to quantify or operationalize these constructs,” Parrish said. “For us, it was important to take a broad purview in this research to see where we did see evidence of improvement in these metrics, and maybe ones where we don’t.”

Ambient AI use also was not associated with an immediate decline in after-hours documentation time, but a statistically significant sustained decline in minutes spent documenting after hours was observed.

"There was an initial decrease in the time spent in notes during the workday, but we didn't see that immediate decline in hours post-workday,” said Robyn Husa, Ph.D., senior clinical research analyst at Providence’s Healthcare Research Accelerator. “We suspect that's because as clinicians were thinking that the AI wrote the notes during the workday, so now I’ve got to go review them, and they were spending more time in that review period, but as it integrates more into their workflow over time, they spend less and less time reviewing it outside of their work hours. It’s more of a testament to the benefits of the gradual increase in use, and the integration into the workflow.”

There were no associations between ambient AI use and appointments per day, but there was an immediate increase in mean RVUs following active ambient AI use.

“We saw an immediate increase of about seven Relative Value Units,” Husa said. “Generally, higher RVUs equate to more services such as labs, imaging, referrals, and follow-ups. Some patients are more complex in their healthcare needs, and they require more time for these services. So our finding there suggests that clinicians who use the ambient AI scribe can perhaps see patients like that more efficiently, or handle these more complex cases, translating to more services. However, there was no change in patient volume per day, meaning clinicians weren't being pushed to see more patients, so they were just spending less time in the documentation, allowing for the billing of more services for each of these patients. I do want to admit that one worry about the introduction of this type of technology is that the system might punish providers for being more efficient by increasing the number of patients they see and we did not see that happen here. It just allowed them to be with their patients more.”