Study Tracks 270 Million Clinical AI Queries on OpenEvidence

A study led by researchers at Harvard Medical School and Brigham and Women's Hospital examined how 950,000 US clinicians used OpenEvidence, according to Harvard Medical School. The study covered January 2024 to June 2026 and was published in JAMA on October 15.

Monthly active users increased from 355 to 703,265 during the study period. Physicians, physician assistants, and nurse practitioners submitted 270 million queries. Researchers used a large language model to classify a random sample of 38.8 million queries.

Clinical decision making accounted for 51.9 percent of queries, with many focused on treatment and patient management. Diagnosis accounted for 13.9 percent. Clinicians also frequently asked about subjects outside their specialties.

Physicians made up more than half of users and submitted an average of 37 queries per month of use. Internal medicine, oncology, and hematology generated a larger share of queries than their share of the physician workforce. Three study coauthors were OpenEvidence employees, and the company provided the data but did not fund the research.