Warning: This is an author submission that is not peer-reviewed or edited. Preprints - unless they show as "accepted" - should not be relied on to guide clinical practice or health-related behavior and should not be reported in news media as established information.
Documenting ICU Rounds with an Ambient Artificial Intelligence Scribe: A Mixed-Methods Study
ABSTRACT
Background:
Ambient artificial intelligence (AI) scribes are increasingly used to support clinical documentation, but their application in intensive care units (ICUs) remains understudied. The complexity of multidisciplinary rounds and ICU documentation may introduce unique challenges to integrating these tools into clinical workflows. Little is known about how ambient AI scribes affect ICU physicians’ workload, documentation practices, and clinical tasks.
Objective:
To characterize ICU attending physicians’ experience using an ambient artificial intelligence (AI) scribe during multidisciplinary rounds and note creation, including perceived impacts, implementation outcomes, and changes in documentation work.
Methods:
We conducted a prospective convergent mixed-methods pilot study at two hospitals in one academic health system. Fifteen ICU attendings completed surveys before and after a 7-day ICU service block using an ambient AI scribe; 11 completed semi-structured interviews. We summarized quantitative measures descriptively and analyzed interviews using rapid qualitative analysis and interview-derived task analysis, followed by mixed-methods integration.
Results:
Mean task-load ratings decreased from pre- to post-pilot for mental demand (72.1 [SD 26.3] to 44.5 [SD 21.1]), physical demand (51.7 [SD 25.4] to 30.8 [SD 22.1]), time pressure (73.1 [SD 23.5] to 41.7 [SD 26.6]), and effort (77.0 [SD 14.3] to 48.2 [SD 26.9]). Mean work-exhaustion scores decreased from 1.60 (SD 0.69) pre-pilot to 0.88 (SD 0.64) post-pilot. Twelve participants (80%) rated the scribe acceptable, appropriate, and feasible. Nine (60%) perceived improved documentation quality; all reported editing generated content. Eight participants reported saving documentation time; six reported increased documentation time. Task analysis identified 17 attending tasks that were new, modified, reduced, and largely unchanged. At 3 months, 11 participants (73%) continued using the scribe.
Conclusions:
ICU attending experiences were generally favorable, but perceived documentation effects varied and scribe use redistributed documentation work. These findings suggest ICU ambient AI scribe implementation should account for workflow fit, verification work, and documentation-task requirements in addition to usability and time.
Citation
Request queued. Please wait while the file is being generated. It may take some time.
Copyright
© The authors. All rights reserved. This is a privileged document currently under peer-review/community review (or an accepted/rejected manuscript). Authors have provided JMIR Publications with an exclusive license to publish this preprint on it's website for review and ahead-of-print citation purposes only. While the final peer-reviewed paper may be licensed under a cc-by license on publication, at this stage authors and publisher expressively prohibit redistribution of this draft paper other than for review purposes.