Banner

News

Article

AI cuts physician time and frustration with EHRs in some instances

Study examines five-week period when family, internal medicine, pediatrics physicians work with artificial intelligence program to log patient notes.

medical technology ehr: © tippapatt - stock.adobe.com

© tippapatt - stock.adobe.com

Artificial intelligence (AI) programs have potential to help physicians and other clinicians reduce the amount of time spent logging patient information in electronic health records (EHRs).

A new study found family medicine, internal medicine and general pediatrics doctors and advanced practice practitioners reported spending less time on EHRs when trained on and aided by an AI-powered clinical documentation tool.

Researchers stated the potential benefits of using AI in health care “to reduce physician burnout, optimize workflows, and refine the accuracy of clinician documentation.” The AI assistant is a program that can listen to physician and patient conversations and use that as the basis to generate a preliminary clinical note.

The researchers, working with Wake Forest University School of Medicine and Atrium Health outpatient clinics in North Carolina and Georgia recruited 85 participants to work with the AI program, which was Dragon Ambient eXperience (DAX) Copilot (Nuance). Another 55 clinicians served as a control group.

Participants answered a seven-question survey before using the AI tool, and five weeks after. They used a numerical rating system to score their experience with it.

After working with the AI program, the intervention group reported:

  • 47.1% (40 of 85) had decreased time in the EHRs at home, compared with 14.5% (eight of 55) in the control group.
  • 44.7% (38 of 85) had decreased weekly time on the EHR outside normal work hours, compared with 20% (11 of 55) in the control group.
  • 43.5% (37 of 85) had decreased time on documentation after patient visits, compared with 18.2% (10 of 55) in the control group.
  • 44.7% (38 of 85) had less frustration using the EHR, compared with 14.5% (eight of 55) in the control group.

A mean of 44.7% of participants in the intervention group, and 68.7% of clinicians in the control group, reported EHR experiences comparable before and after the experiment. The researchers noted “a significant subset did not find time-saving benefits or improved EHR experience.” They also said selection and recall bias could have affected both groups, limiting the study.

The research letter, “AI-Powered Clinical Documentation and Clinicians’ Electronic Health Record Experience: A Nonrandomized Clinical Trial,” was published in JAMA Network Open.

Related Videos