Atrium Health DAX Copilot Study Finds 7% Documentation Time Reduction Only for High-Utilization Clinicians

Atrium Health deployed Large Language Models & Generative AI for Clinical Documentation & Patient Records in Hospital & Health System. As reported by www.healthcareitnews.com: 7% documentation hours reduction (high users).

Maintained by Peter Korpak, Lead EditorHow evidence is checked
7%Documentation Hours Reduction (High Users)
75% (84/112)Active Users Transferring >25% of Notes to Epic
~60% (67/112)High Users Transferring >60% of Notes to Epic

Source-reported figures (cited source: www.healthcareitnews.com)

What Atrium Health was trying to fix

Atrium Health sought to reduce clinician documentation burden and EHR-related burnout by deploying ambient AI clinical documentation at scale. As one of the first U.S. health systems to implement such technology, the organization needed to determine whether broad rollout of AI-assisted documentation would yield meaningful efficiency gains across its outpatient clinician workforce.

What Atrium Health deployed

Atrium Health deployed Nuance's DAX Copilot across outpatient clinics in North Carolina and Georgia, integrated with Epic EHR. A total of 238 clinicians in family medicine, internal medicine, and general pediatrics were enrolled across five waves between June and August 2023, with DAX users receiving one-hour training and EHR account setup. Researchers from Wake Forest University School of Medicine evaluated outcomes over 180 days.

Results

The NEJM AI-published study found few statistically significant differences between DAX users and the control group across key metrics including time in EHR, work outside work hours, note completion rates, and financial metrics. High DAX users — those who transferred more than 60% of notes to Epic — showed approximately 7% fewer documentation hours compared to controls. Researchers concluded that widespread implementation in its current form is unlikely to generate appreciable productivity gains for healthcare systems broadly.

Key Takeaways

  • High utilization threshold (>60% of notes transferred) appears necessary to realize meaningful documentation time savings, suggesting adoption depth matters more than adoption breadth
  • Ambient AI documentation may benefit select clinician subgroups rather than serving as a universal efficiency solution across all specialties and practice styles
  • Broad rollout without targeting high-adoption clinicians is unlikely to justify AI documentation investment on productivity grounds alone

Evidence for Atrium Health's Clinical Documentation & Patient Records deployment

Reported outcome metrics
3 cited below
Last updated
Source link checked

Share:

Have a similar implementation?

Share your customer's AI results and link it to your vendor profile.

Submit a case study →