Evaluating an Ambient Artificial Intelligence Scribe for Documentation Quality and Efficiency in Psychiatric Consultations: A Simulation-based Study

Background Documentation demands in psychiatric practice diminish time for direct patient care and are associated with clinician burnout. Ambient artificial intelligence (AI) scribes may facilitate more efficient and higher-quality documentation while reducing clinician workload and preserving the integrity of the clinical encounter. This study aims to determine the impact of an ambient AI scribe in improving documentation quality and efficiency while reducing clinician workload during simulated psychiatric consultations.

We conducted a prospective, cross-over, within-subject simulation study to compare conventional keyboard-based documentation (traditional condition) with documentation assisted by an ambient artificial-intelligence scribe (AI-scribe condition). The study was conducted at an academic simulation centre comprising of eight clinicians with psychiatric experience who completed both documentation conditions. Clinician workload (NASA Task Load Index [NASA-TLX]), documentation quality (Sheffield Assessment Instrument for Letters [SAIL]), video-verified screen time during documentation, and bespoke clinician and patient experience questionnaires were administered.

AI-scribe use was associated with substantially lower workload versus traditional documentation (NASA-TLX total, 25.0 versus 461.3; mean difference, 436.25; 95% CI, 389.93-482.57; p<0.001), with significant improvements in 5 of 6 subscales, including temporal demand (mean difference, 50.00; p<0.001), frustration (mean difference, 33.75; p=0.04), and perceived performance (mean difference, 30.0; p=0.03). Documentation quality improved with AI-scribe (SAIL; 22.88 versus 14.38; t(7)=2.55; p=0.038; Cohen’s dz=0.90); checklist completeness was ≥75% on 17 of 20 items, with 9 items at 100%. Screen time decreased (7.46 versus 5.31 minutes; mean difference, −2.15; 95% CI, −2.10-+6.41; p=0.27; Hedges g=0.38); 5 of 8 clinicians showed individual reductions. Overall clinician satisfaction was higher with AI-scribe (100% versus 50%), and 87.5% agreed that AI assistance reduced cognitive load. Patient-reported experience favoured the AI-scribe condition.

Ambient AI scribe can assist in improving documentation quality and substantially reducing clinician workload while maintaining favourable patient-perceived consultation quality. Competing Interest Statement The authors have declared no competing interest. Funding Statement This study received funding from Lyrebird Health who, has no role in study design, data collection, data interpretation or manuscript writing. None of the authors or research participants are affiliated with Lyrebird Health. Author Declarations I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained. Yes The details of the IRB/oversight body that provided approval or exemption for the research described are given below: All procedures involving human subjects/patients were approved by the Dubai Scientific Research Ethics Committee (DSREC) waived formal review (MBRU IRB-2025-203). I confirm that all necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived, and that any patient/participant/sample identifiers included were not known to anyone (e.g., hospital staff, patients or participants themselves) outside the research group so cannot be used to identify individuals. Yes I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance). Yes I have followed all appropriate research reporting guidelines, such as any relevant EQUATOR Network research reporting checklist(s) and other pertinent material, if applicable. Yes Data Availability Data collected and analysed , including an research materials, will be readily shared upon direct request by email to the corresponding author of this study.