Study Describes Clinician Use of AI Platform OpenEvidence

A JAMA cross-sectional study characterized use of OpenEvidence, an artificial intelligence–enabled clinical information platform, among clinicians with National Provider Identifier–verified accounts. The analysis covered January 2024 through June 2026 and included 952,008 clinicians, including 556,200 physicians. A large language model classified the clinical purpose and topics of a stratified random sample of 38.8 million queries.

Clinicians submitted 271 million queries overall. Monthly active users increased from 355 to 703,265, while average monthly queries per active user rose from 18.0 to 39.5. Nurse practitioners and physician assistants each submitted 44.5 queries per active user-month, compared with 37.1 among physicians. Relative to national workforce representation, query use was higher among physicians within a decade of graduation, residents and fellows, and physicians in oncology and hematology, internal medicine, and family medicine.

Clinical decision-making accounted for 51.9% of queries, including treatment and management at 33.1%; diagnosis accounted for 13.9%. Queries outside physicians’ own specialties ranged from 35% among oncologists and hematologists to 88% among emergency department physicians. These findings describe information-seeking behavior, not the accuracy of platform responses or effects on patient care. Interpretation is limited by the single-platform, cross-sectional design and automated query classification; the supplied abstract does not report classification validation or funding disclosures.