Early insights into the impact of an ambient AI scribe solution on clinical documentation to reduce clinician burnout in oncology.

e13722 Background: Documentation burden can cause clinician burnout leading to poor decision making, patient safety events and clinician turnover. Innovations in ambient artificial intelligence (AI) hold promise to minimize burden. Ambient AI scribes use voice recognition technology to securely listen to a clinician-patient interaction and leverage generative AI to summarize that interaction for review and editing as a note. Our understanding of ambient AI solutions in oncology is not well understood but necessary because of the complexity of oncology visits. Our objectives were to evaluate the impact of the technology on clinician documentation and the potential to reduce burden, and implementation outcomes in oncology. Methods: This prospective pilot study was conducted with providers across specialties over 4 months at a comprehensive cancer center. Post-pilot surveys and interviews were used to collect implementation outcomes from the Proctor framework including adoption, feasibility, acceptability, and appropriateness. The analysis included utilization outcomes, descriptive statistics of survey data, and content analysis from interviews. Results: Thirty-one providers were recruited and 25 used the application at least once; 18 completed a survey and 21 an interview. Providers used the technology in 620 (14%) out of 4449 outpatient visits. Of 18 survey respondents, 17 (94%) indicated they used the AI-drafted content at least sometimes, demonstrating feasibility. For acceptability, 11 (61%) indicated a moderate/strong desire for continued access. For burden, 44% indicated that using the application reduced time spent outside clinic hours, whereas 17% indicated an increase. Visit type was one source of variation in appropriateness, with utilization higher for new visits compared to follow-ups (21% vs. 12%). From interviews, some stated the technology provided value across clinical scenarios and patients; others expressed concerns related to capturing complex clinical scenarios and trial details. Some noted adoption barriers due to lack of EHR integration. All advocated for continued investment. Conclusions: Clinicians generally found the technology to be feasible, acceptable, and appropriate, with potential to decrease documentation burden. Future research should evaluate this technology at scale to realize workflow efficiencies, and its capacity to meaningfully improve provider satisfaction and reduce burnout.