Ann Intern Med

AI excels in virtual urgent care decision support

April 15, 2025

Study details: This retrospective cohort study analyzed 461 physician-managed visits at Cedars-Sinai Connect, an AI-assisted virtual urgent care clinic, from June 12 to July 14, 2024. The study compared initial AI recommendations with final physician recommendations for adults presenting with respiratory, urinary, vaginal, eye, or dental symptoms.

Results: AI and physician recommendations were concordant for 262 (56.8%) visits. AI recommendations were rated as optimal in 77.1% of cases vs. 67.1% for physician decisions. Quality scores were equal in 67.9% of cases, better for AI in 20.8%, and better for physicians in 11.3%.

Clinical impact: These findings suggest that AI can enhance decision-making in virtual urgent care settings, particularly in identifying critical issues and adhering to guidelines. Integrating AI into clinical practice may improve patient outcomes and streamline care delivery.

Source:

Zeltzer D, et al. (2025, April 4). Ann Intern Med. Comparison of Initial Artificial Intelligence (AI) and Final Physician Recommendations in AI-Assisted Virtual Urgent Care Visits. https://pubmed.ncbi.nlm.nih.gov/40183679/