Artificial Intelligence Scribes Improve Clinician Experience and Efficiency Among Pediatric Subspecialists

Strategic Insights -

Researchers at the Children's Healthcare of Atlanta Inc. conducted a randomized quality improvement trial to examine the impact of artificial intelligence (AI) scribes on clinician experience and efficiency among pediatric subspecialists. The trial involved 22 providers who used AI scribes to generate provider note sections from recordings of patient-provider encounters. The results showed that AI scribes improved clinician experience and burnout without increasing charting time or EHR work outside work hours.

Key Takeaways:

  • The trial involved 22 providers, with 12 in the intervention group and 10 in the control group.
  • AI scribes were used to populate 1 character in 53% of visit notes, with a range of 10.6% to 98.2% across providers.
  • The KLAS Net EHR Experience survey improved among intervention providers from 52.6 to 75.2, while dropping from 37.3 to 30 among control providers.
  • Experiencing burnout dropped from 8 (89%) to 5 (56%) among intervention providers, while remaining stable at 3 (43%) in the control group.
  • There was no significant change to pajama time (-9.4 minutes per scheduled day), time in notes per encounter (+0.2 minutes per note), or work Relative Value Units (wRVUs) per encounter (-0.03).
  • Of 48 qualitative comments, 69% had a positive sentiment, 15% neutral, and 17% negative.

Statistics:

  • The trial involved 22 providers, with 12 in the intervention group and 10 in the control group.
  • AI scribes were used to populate 1 character in 53% of visit notes, with a range of 10.6% to 98.2% across providers.
  • The KLAS Net EHR Experience survey improved by 22.6 points among intervention providers, while dropping by 7.3 points among control providers.
  • Experiencing burnout decreased by 3 (89%) points among intervention providers, while remaining stable at 0 points in the control group.
  • There was no significant change to pajama time (-9.4 minutes per scheduled day), time in notes per encounter (+0.2 minutes per note), or work Relative Value Units (wRVUs) per encounter (-0.03).

Sources:

  • NewsRx. Data on Artificial Intelligence Reported by Nikolay P. Braykov and Colleagues (The Influence of Artificial Intelligence Scribes on Clinician Experience and Efficiency Among Pediatric Subspecialists: A Rapid Randomized Quality Improvement Trial). Pediatrics Week. August 2, 2025; p 162.
  • The Influence of Artificial Intelligence Scribes on Clinician Experience and Efficiency Among Pediatric Subspecialists: A Rapid Randomized Quality Improvement Trial. Applied Clinical Informatics, 2025.