TL;DR
Turn your wind-down time into listening time
- Thousands of audiobooks, podcasts and originals
- Listen on your phone, tablet or Echo — also offline
- Cancel anytime
A study of an AI ambient scribe at Beth Israel Lahey Health found that clinicians who used it spent less time actively working on notes, while their notes became longer. Same-day and after-hours note closure and clinical productivity did not improve; researchers and an accompanying editorialist said further study is needed.
A study of an AI ambient scribe used in outpatient care across a 14-hospital system found that clinicians spent less time actively working on notes after enrolling, but did not improve same-day or after-hours note closure or clinical productivity. The findings, reported by Jennifer P. Stevens of Beth Israel Deaconess Medical Center and colleagues in Annals of Internal Medicine, show a limited documentation-time benefit rather than a broader productivity gain.
Among clinicians who used the Heidi system, median note-active time fell from 21 minutes to 18.6 minutes after enrollment. Median note length rose from 4,854 characters to 5,187 characters. The researchers reported no corresponding difference in same-day or after-hours note closure, and no improvement in measures of clinical productivity.
The study examined Heidi’s introduction at Beth Israel Lahey Health, where outpatient clinicians could choose whether to enroll and use the tool separately from the electronic health record. Of 4,993 clinicians providing direct patient care in 2025, 1,140 enrolled between January and November. After exclusions, 902 clinicians were included in the analysis; 67% used Heidi at least once, while 33% did not use it after enrolling.
Effects varied by use level. In the first three months after enrollment, high users had a larger immediate reduction in note-active time—2.50 minutes—than enrolled nonusers, at 0.29 minutes, or low users, at 0.78 minutes. The reduction for low users did not differ significantly from that for enrolled nonusers. The results suggest that any time benefit was concentrated among clinicians who used the tool more often.
Time Savings Without Wider Gains
The findings address a practical question for health systems considering AI documentation tools: whether less time spent on notes translates into more completed work or more time for patients. In this study, the measured reduction in note-active time did not coincide with better note-closure timing or clinical productivity. A narrower time measure should not be treated as evidence that overall workload fell.
Documentation is also linked to clinicians’ working conditions. The study authors noted that 43.2% of clinicians reported burnout symptoms in 2024, and cited electronic health record work as a contributor. The results offer some evidence that frequent use may reduce one part of documentation work, but they do not establish that the tool reduces burnout, changes patient care, or saves time across the full clinical workflow.
Popular right now
Sony WH-1000XM4C Wireless Noise-Cancelling Headphones, Black
Ninja Foodi Air Fryer 10 Qt DualZone 6-in-1 Black DZ302
Ninja DualZone Smart Air Fryer, 10 Qt, 6-in-1, DZ950
As an affiliate, we earn on qualifying purchases.
How Heidi Was Studied
Beth Israel Lahey Health introduced Heidi in phases beginning in January 2025. Physicians, non-physician practitioners, trainees and allied health professionals could enroll voluntarily. The research compared outcomes before and after enrollment among participating clinicians, and also examined differences between high users, low users and people who enrolled but did not use the system.
The study focused on outpatient clinicians with at least 10 encounters per month, then excluded some clinicians with too few encounters or too little time in the study periods. It was an evaluation of one tool in one health system, not a comparison of every ambient scribe product. The authors said future research should examine productivity, documentation accuracy and health equity.
“By reducing documentation burden, AI scribes offer a promising strategy to address a contributor to EHR-related clinician burnout.”
— Jennifer P. Stevens and colleagues, study authors
Unmeasured Editing and Note Quality
The study did not measure how long clinicians spent editing notes within Heidi before transferring them to the EHR. Ostherr said that gap makes it difficult to determine whether the system reduced documentation work overall or shifted part of it to another step. The increase in note length also remains unexplained, and the study does not establish what that change means for care quality.
The analysis does not show whether AI-generated notes contained errors, omissions or misinterpretations, nor does it establish effects on patient outcomes or burnout. The editorialist said the long-term impact of incorporating AI-generated text into clinical notes is unknown. Findings from a single health system also may not apply to other tools, workflows or clinician groups.
Further Evaluation of Clinical Use
The study authors called for additional research into clinical productivity, documentation accuracy and health equity. Future evaluations could track time spent reviewing and editing generated notes, as well as whether changes in note length affect the information clinicians record and use. The supplied report does not identify a scheduled follow-up study or a specific next publication date.
For now, the findings support a measured conclusion: frequent Heidi use was associated with a greater reduction in note-active time, but the study did not find broader productivity gains. Whether the tool reduces total documentation burden, improves working conditions or affects the quality of clinical records remains unsettled.
Key Questions
What did the study find about AI scribe use?
After enrollment in Heidi, median note-active time among users decreased from 21 minutes to 18.6 minutes, while median note length increased. Researchers found no improvement in same-day or after-hours note closure or clinical productivity.
Which clinicians saw the largest time reduction?
High users had an immediate reduction of 2.50 minutes in note-active time during their first three months, compared with 0.29 minutes for enrolled nonusers and 0.78 minutes for low users. The low-user result did not differ significantly from the nonuser result.
Did the study show that AI scribes reduce burnout?
No. The study assessed documentation and productivity measures, not whether use reduced clinician burnout. The authors described AI scribes as a potential way to address a contributor to burnout, while calling for more research.
What is still unknown about the results?
The study did not capture time clinicians spent editing notes inside Heidi before moving them to the EHR. It also did not establish why notes became longer or what that meant for documentation accuracy, patient care or the overall workload.
Source: medpagetoday.com
