Clinicians Use AI Platform More for Treatment Questions Than Diagnosis

A total of 952,008 US clinicians submitted 271 million queries to an artificial intelligence (AI)–enabled clinical information platform over 2.5 years, and those queries more often concerned treatment and management than diagnosis, according to a cross-sectional study published in JAMA.

Investigators analyzed query logs from January 1, 2024, through June 30, 2026, for verified clinician accounts on OpenEvidence, a platform that uses large language models (LLMs) to generate referenced answers to clinical questions. The authors described it as the most commonly used platform of its kind in the US and noted that empirical evidence on how clinicians use these tools remains limited.

Physicians accounted for 556,200 users and 172.1 million queries. Monthly active users rose from 355 in January 2024 to 703,265 in June 2026. In a stratified random sample of 38.8 million queries classified by an LLM, clinical decision-making made up 51.9% of physician queries; treatment and management were the most frequent purpose (33.1%), and diagnosis accounted for 13.9%.

Physicians within a decade of graduation submitted a larger share of queries than their share of the workforce (use ratio, 2.54), as did residents and fellows (1.84) and physicians in oncology and hematology (2.00), internal medicine (1.88), and family medicine (1.80). Ratios were lowest in procedural and diagnostic specialties such as ophthalmology (0.23) and radiology (0.24). Queries outside physicians’ own specialties ranged from 35% among oncologists and hematologists to 88% among emergency department physicians.

The authors noted that the analysis was limited to a single platform, relied on an LLM to classify queries, and could not determine whether queries occurred during patient encounters. Three authors disclosed employment or leadership roles at OpenEvidence, which provided data and contributed to the analysis. The investigators concluded that efforts to evaluate, teach, and integrate AI in clinical care should draw on empirical data on its real-world use.

Source: Patel VR, Finlayson SG, Liu-Hass M, et al. Clinician use of an artificial intelligence–enabled clinical information platform. JAMA. Published online October 15, 2026. doi:10.1001/jama.2026.21868