Surveys Reveal Increased Physician Productivity, Billing From AI Scribe Implementation

The integration of AI ambient scribes into electronic medical record (EMR) systems in oncology clinics may lead to greater satisfaction among physicians regarding workflow and fewer perceived challenges when dealing with EMRs. In addition, findings from a single-center, randomized study published in JCO Oncology Practice show that use of AI scribes increased the mean clinic visits and total billings per working day that each physician could achieve.

Earlier findings from this analysis were presented at the 2025 ASCO Annual Meeting (Abstract 11167).

Medical scribes using AI are increasingly being adopted in outpatient clinical settings to improve workload efficiency and reduce the amount of time spent on patient documentation, thereby allowing for more productive time with patients. However, most scribes have not been evaluated by a regulatory body prior to their adoption, leading to ongoing concerns about quality control, oversight, privacy, and more. Few studies to date have explored the benefits of AI scribes on workload burden reduction and billing.

Investigators conducted a single-center, two-phase, mixed-design study of AI scribe integration with EMRs in a multisite community oncology practice. Twenty-two oncologists and three primary care physicians from Astera Cancer Care were enrolled in the study and received training on the Knowtex AI scribe for use during clinic visits. Randomization allowed some physicians to receive earlier access to the AI scribe (week 0 vs 16). The study was conducted as part of the onboarding process for the Knowtex AI scribe into the practice’s OncoEMR system. During the testing periods, the AI scribe was used for more than 90% of all patient visits.

The participating physicians answered survey questions about their motivation, work-life balance, documentation quality, and EMR challenges both at baseline and after 8 weeks.

The primary outcome measures for productivity were the mean number of visits per working day and the mean total fees per working day. Secondary outcome measures included diagnostic codes per visit and mean fee per visit to determine any changes in billing complexity or visit intensity.

Physicians in the intervention cohort were generally younger and demonstrated a greater understanding of and motivation to use AI scribes compared with those in the control cohort.

In the intervention cohort that received earlier access, use of the AI scribe led to an increase in the mean number of patient visits per day from 16.63 to 18.45 (P < .01), and the mean total fees per day grew from $7,101.60 to $7,893.70 (P < .01). However, there were no statistically significant differences in either measurement in the control cohort. Fees per visit, diagnostic codes per visit, and delays in bill posting did not show significant differences with or without the AI scribe in the intervention cohort.

Only 76% of the physicians completed the baseline survey and 64% completed the follow-up survey. The belief that documentation impacted work–life balance negatively and the amount of time spent on documentation both decreased with the use of the AI scribe.

The survey also revealed that physicians spent more time interacting with patients during their visits post–AI scribe exposure (P = .011) and less time documenting the visit within the EMR system (P = .003). Additionally, physicians were more satisfied with the process of EMR documentation (P = .003) and reported fewer challenges with EMR (P < .001) following AI scribe implementation. Opinions did not change significantly, however, with regard to documentation quality.

“AI scribe implementation with EMR integration increased daily patient visits without diminishing physician-reported work-life balance or documentation satisfaction, aligning with reported benefits of traditional scribes. Physicians reported less in-clinic documentation time, improved workflow satisfaction, and fewer EMR-related challenges,” the study authors, led by Nima Toussi, BSc, of the College of Medicine, University of Saskatchewan, Saskatoon, Canada, wrote in their published report. “Although fee per visit remained unchanged, overall volume-based gains in productivity suggest that AI scribes can streamline documentation workflows.”

DISCLOSURE: Ms. Toussi reported an uncompensated relationship with Knowtex Inc. Ms. Zhang and Ms. Kang are both employed by Knowtex Inc. Dr. Licitra is employed by Astera Cancer Care and reported stock and ownership interests in AbbVie, Exelixis, Lilly, and Vertex. No other potential conflicts of interest were reported.

AI in Practice: Does your practice use AI scribes? Have you benefited from using AI scribes or been challenged by their implementation? Share your experiences with the oncology community.

ASCO AI in Oncology is published by Conexiant under a license arrangement with the American Society of Clinical Oncology, Inc. (ASCO^®). The ideas and opinions expressed in ASCO AI in Oncology do not necessarily reflect those of Conexiant or ASCO. For more information, see Policies.