Models · OpenAI
Study of 40,000 patient visits finds clinicians using AI copilot made fewer errors
Compiled by KHAO Editorial — aggregated from 1 outlet. See llms.txt for citation guidance.
★ Tier-1 Source
AI systems have the potential to improve human health globally—to make reliable health information universally available, help clinicians deliver better care, and empower people to better understand and advocate for their health.
Key facts
- The study analyzed data from 39,849 patient visits: 20,859 in the group with AI Consult (the AI group ) and 18,990 in the group without (the non-AI group )
- 108 independent physicians (29 from Kenya) rated the final documentation and decisions from 5666 randomly selected visits to identify errors
- History-taking errors were reduced by 32%, investigations errors by 10%, diagnostic errors by 16%, and treatment errors by 13%
- Significance levels are denoted by stars: for p ≤ 0.05, for p ≤ 0.01, and for p ≤ 0.001
Summary
Study of 40,000 patient visits finds clinicians using AI copilot made fewer errors. Large language model (LLM) performance and safety in health continue to advance. To advance research on real-world implementation, OpenAI partnered with Penda Health , a primary care provider operating in Nairobi, Kenya since 2012, to conduct a novel study of Penda’s LLM-powered clinician copilot. In a study of 39,849 patient visits across 15 clinics, clinicians with AI Consult had a 16% relative reduction in diagnostic errors and a 13% reduction in treatment errors compared to those without.