TOPLINE: 

Artificial intelligence (AI) scribes were associated with reduced burnout among ambulatory care physicians and advanced practice practitioners after 30 days of use. The technology improved cognitive task load and attention to patients and reduced time spent on documentation in US health systems.

METHODOLOGY:

• Researchers conducted a quality improvement study from February 1 to October 31, 2024, to evaluate the effect of using an ambient AI scribe on administrative burden and clinician burnout over 30 days in ambulatory clinics.
• Participating clinicians recruited voluntarily by digital health leaders across six US health systems used an ambient AI scribe for 30 days to assist with clinical documentation.
• Clinicians selected a patient appointment, obtained verbal consent, and recorded the visit using the ambient AI scribe. Documentation was instantly generated online, allowing for viewing, editing, and automatic import into the clinician’s note template.
• Of the recruited participants, 263 (mean years in practice, 15.1; 53.6% women) who completed both preintervention and 30-day postintervention surveys were included. They were comprised of 131 primary care professionals , 46 adult specialists, 14 in neurology and psychiatry, and 72 in surgical specialties.
• The primary outcome was self-reported burnout, assessed using a 5-point metric validated against the emotional exhaustion domain of the Maslach Burnout Scale, with secondary outcomes including changes in cognitive task load, focused patient attention, and time spent on documentation after hours.

TAKEAWAY:

• Among 186 participants, the proportion reporting burnout (assessed using the standard cutoff of ≥ 3) decreased from 51.9% to 38.8%. The odds of clinician burnout decreased by 74% after 30 days of using ambient AI scribes (adjusted odds ratio, 0.26; P < .001).
• Participants experienced improvements in the secondary outcomes note-related cognitive task load and focused attention to patients (P < .001 for both).
• The ability to add patients to the clinic schedule if urgently needed improved (P = .02), and the time spent documenting after hours decreased (P < .001).
• Subgroup analyses revealed significant reductions in burnout among medical doctors (P < .001) and both academic (P = .01) and medical group-employed (P = .001) clinicians. Significant reductions in burnout across ambulatory specialties — including family medicine and pediatrics, obstetrics and gynecology, and adult specialties — were observed.

IN PRACTICE:

“Artificial intelligence scribes may represent a scalable solution to reduce administrative burdens for clinicians and allow more time for meaningful work and professional well-being. Ambient Al solutions may be scalable at a lower cost than human scribes,” the authors wrote.

SOURCE:

This study was led by Kristine D. Olson, MD, MSc, Yale School of Medicine in New Haven, Connecticut. It was published online on October 02, 2025, in JAMA Network Open.

LIMITATIONS: 

As a quality improvement initiative, the evaluation was not structured for research purposes. The study could not adjust for temporal trends as a control group was lacking. The dataset only included complete survey responses, preventing characterization of noncompleters. Findings were based on subjective reports without quantitative data from electronic health records.

DISCLOSURES:

The Abridge ambient AI intervention was funded by the participating institutions. One author disclosed being a senior director of clinical success at Abridge AI. Another author reported being the chief clinical officer of Abridge AI during the study’s conduct, and a third author reported serving on a research advisory committee for Abridge AI. One author reported receiving personal fees from Intrinsic Brands.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.