Reporting completeness of clinical trials evaluating large language model interventions in medicine: systematic review.

Reporting completeness of clinical trials evaluating large language model interventions in medicine is modest, with average adherence around 40-61% across various guidelines.

• Why it matters: Incomplete reporting hampers the ability to assess trial quality, reproducibility, and safety, which are critical for translating LLM interventions into clinical practice.
• What they did: A systematic review and meta-research analyzed 68 trial reports, 15 protocols, and 378 registrations from 2017 to 2025, assessing adherence to established reporting guidelines.
• The result: Findings reveal persistent underreporting of methodological details, reproducibility artifacts, and harms, limiting confidence in the safety and effectiveness of LLM interventions in healthcare.