Randomized Trial Protocol: Head-to-Head Comparison of Two Ambient Artificial Intelligence Documentation Tools on Clinician Usability
- 1UCLA Health Information Technology, UCLA Health, University of California, Los Angeles
Abstract
Background: Ambient AI documentation reduces documentation burden and electronic health record (EHR) time for ambulatory clinicians. Newer features also queue orders and offer some level of coding support for clinician review. This study compares two commercial products deployed at UCLA Health. Objective: To compare Nabla and Epic Systems, Inc. ambient AI solution among ambulatory clinicians at UCLA Health on usability, documentation burden, perceived note quality, adoption, EHR time and productivity, clinician-reported experience with order queueing and coding support, coding and billing patterns and integrity, work productivity, and patient-reported communication experience. Methods: Two-arm parallel-group randomized trial, 1:1, among 374 ambulatory clinicians at a single academic health system who hold an active license for a standalone ambient application (Nabla). Randomization is covariate-constrained where cohort size permits, balancing age, sex, half-day clinic sessions, specialty group, and baseline System Usability Scale (SUS) score. Intervention runs 11/9/2026 to 1/17/2027, with the first four weeks as lead-in. The primary outcome is SUS score at follow-up, compared between arms. Analysis is by intention-to-treat using mixed-effects models. Ethics and dissemination: Determined exempt by the UCLA Institutional Review Board (IRB-26-2147). Survey completion indicates consent. No patient-level data are collected. Findings will directly inform operational decision making and will be published regardless of direction.
Related
- A prebill control plane for record integrity in inpatient artificial intelligence-enabled clinical documentation workflows.
- A Framework to Monitor Editing of Artificial Intelligence-Generated Medical Documentation
- NOTIFY-DISCUSS: Patient and clinician perspectives on opportunistic screening for coronary artery calcification using artificial intelligence
- Artificial intelligence in medicine: a clinician-oriented guide and operational framework to evaluate new AI devices
- The Causal Artificial Intelligence Clinician for early haemodynamic management of septic shock in ICU
BibTeX
@misc{chin2026randomized,
title = {{Randomized Trial Protocol: Head-to-Head Comparison of Two Ambient Artificial Intelligence Documentation Tools on Clinician Usability}},
author = {A. T. Chin and K. Leung and S. Vangala and H. Woo and T. Kingsley and A. Sandhu and L. Jalalian and E. M. Cheng and J. N. Mafi and P. J. Lukac},
howpublished = {medRxiv},
year = {2026},
month = nov,
doi = {10.64898/2026.11.07.26367514},
url = {https://www.medrxiv.org/content/10.64898/2026.11.07.26367514v1},
note = {CC BY-NC-ND},
}
RIS
TY - UNPB
TI - Randomized Trial Protocol: Head-to-Head Comparison of Two Ambient Artificial Intelligence Documentation Tools on Clinician Usability
AU - A. T. Chin
AU - K. Leung
AU - S. Vangala
AU - H. Woo
AU - T. Kingsley
AU - A. Sandhu
AU - L. Jalalian
AU - E. M. Cheng
AU - J. N. Mafi
AU - P. J. Lukac
PY - 2026
DA - 2026/11/07
PB - medRxiv
DO - 10.64898/2026.11.07.26367514
UR - https://www.medrxiv.org/content/10.64898/2026.11.07.26367514v1
AB - Background: Ambient AI documentation reduces documentation burden and electronic health record (EHR) time for ambulatory clinicians. Newer features also queue orders and offer some level of coding support for clinician review. This study compares two commercial products deployed at UCLA Health. Objective: To compare Nabla and Epic Systems, Inc. ambient AI solution among ambulatory clinicians at UCLA Health on usability, documentation burden, perceived note quality, adoption, EHR time and productivity, clinician-reported experience with order queueing and coding support, coding and billing patterns and integrity, work productivity, and patient-reported communication experience. Methods: Two-arm parallel-group randomized trial, 1:1, among 374 ambulatory clinicians at a single academic health system who hold an active license for a standalone ambient application (Nabla). Randomization is covariate-constrained where cohort size permits, balancing age, sex, half-day clinic sessions, specialty group, and baseline System Usability Scale (SUS) score. Intervention runs 11/9/2026 to 1/17/2027, with the first four weeks as lead-in. The primary outcome is SUS score at follow-up, compared between arms. Analysis is by intention-to-treat using mixed-effects models. Ethics and dissemination: Determined exempt by the UCLA Institutional Review Board (IRB-26-2147). Survey completion indicates consent. No patient-level data are collected. Findings will directly inform operational decision making and will be published regardless of direction.
ER -