Efficacy of a large language model (LLM) data extraction system for simplifying evidence reviews of emerging infectious diseases: A randomized crossover trial

Efficacy of a large language model (LLM) data extraction system for simplifying evidence reviews of emerging infectious diseases: A randomized crossover trial - LLM data extraction system crossover RCT for emerging infectious disease evidence review

Status

Active, not recruiting

Phases

Unknown

Study type

Interventional

Source

JPRN

Registry ID

JPRN-UMIN000058346

Enrollment

Registered

2025-07-02

Start date

2025-07-01

Completion date

Unknown

Last updated

2026-06-29

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Emerging Infectious Diseases (Evidence Review Efficiency Research)

Interventions

LLM System Use (Using structured data extraction system from papers with OpenAI o4-mini latest version to input into Streamlit web form) No LLM System Use (Reading paper PDFs directly and manually inp

Eligibility

Sex/Gender

All

Inclusion criteria

Inclusion criteria: 1. Researchers or healthcare professionals with experience in evidence review 2. Researchers or healthcare professionals specializing in infectious diseases 3. Individuals with basic computer skills 4. Individuals who have consented to participate in the research

Exclusion criteria

Exclusion criteria: 1. Individuals with severe visual impairment 2. Individuals participating in an environment with unstable internet connection

Design outcomes

Primary

Measure	Time frame
Time from authentication of subject ID after accessing the web form and starting input until completion of input	—

Secondary

Measure	Time frame
Accuracy of data extraction (experts check data extracted by participants for errors and calculate the percentage of correct answers)	—

Countries

Japan

Contacts

Public ContactMasahiro Ishikane

National Center for Global Health and Medicine Disease Control and Prevention Center

ishikanemasahiro@gmail.com03-3202-7181

Outcome results

None listed

Source: JPRN (via WHO ICTRP) · Data processed: Jul 3, 2026