AI Outperforms Physicians in Virtual Urgent Care Study, Highlighting Potential for Improved Patient Care

AI Demonstrates Superior Performance in Virtual Urgent Care

A groundbreaking study conducted by Cedars-Sinai has revealed that artificial intelligence (AI) may outperform human physicians in providing treatment recommendations for patients in virtual urgent care settings. The research, presented at the American College of Physicians Internal Medicine Meeting and published in the Annals of Internal Medicine, compared initial AI treatment recommendations to final recommendations made by physicians 1

1

Study Design and Findings

The retrospective study analyzed data from Cedars-Sinai Connect, a virtual primary and urgent care program launched in 2023. Researchers reviewed 461 physician-managed visits with AI recommendations from June 12 to July 14, 2024, focusing on adults with respiratory, urinary, vaginal, vision, or dental symptoms 2

2

Key findings include:

1. AI recommendations were rated higher than final physician recommendations for common complaints in urgent care settings.
2. AI was particularly effective in identifying urinary tract infections potentially caused by antibiotic-resistant bacteria.
3. In about two-thirds of cases, doctors made the same clinical decisions as the AI.
4. In the remaining one-third, AI recommendations were rated superior twice as often as they were inferior 3

   3

   .

Strengths of AI and Human Physicians

The study highlighted distinct strengths of both AI and human physicians:

• AI excelled at identifying critical red flags and following medical guidelines strictly.
• Physicians demonstrated superior skills in eliciting more complete patient histories and adapting recommendations accordingly 4

  4

  .

AI System and Implementation

The AI system used in Cedars-Sinai Connect was developed by K Health, designed to reduce the burdens of clinical intake and data entry. Patients interact with the AI through a mobile app, answering an average of 25 questions in five minutes. The AI then provides initial diagnoses and treatment recommendations, which can be viewed by the treating physician 5

5

Related Stories

AI Diagnosis Outperforms Emergency Room Doctors in Harvard Study, Raises Questions on Patient Care

AI Chatbots Enhance Physician Decision-Making in Clinical Management, Study Finds

AI in Healthcare: Patients Trust AI Medical Advice Over Doctors, Raising Concerns and Challenges

AI Shows Promise in Clinical Decision-Making, But Challenges Remain

Implications and Future Directions

Dr. Caroline Goldzweig, Cedars-Sinai Medical Network chief medical officer and co-senior author of the study, emphasized that AI decision support, when implemented effectively at the point of care, has the potential to improve clinical decision-making for common and acute conditions 1

1

However, researchers noted that the major uncertainty of the study was whether physicians actually viewed and incorporated the AI recommendations into their clinical decision-making 2

2

Expert Commentary

Dr. Jerome Kassirer, a professor at Tufts University School of Medicine, suggested in an accompanying editorial that clinicians should initially use AI programs to confirm their own diagnoses and therapeutic recommendations. He emphasized the importance of exploring discrepancies between AI and human judgments as a means to improve AI tools 5

5