The Diagnostic and Triage Capacity of Laypeople-large Language Model Collaboration in China

NCT07250516 | Completed DMC

• 1 other identifier: JCYJ20240813115806009
• Study Type: interventional
• Target: 6,360
• Locations: 1 country
• Sites: 1

Brief Summary

The goal of this randomized controlled trial is to evaluate the role of large language models in enhancing laypeople's ability to self-diagnose and triage common diseases. The main questions it aims to answer are:

• Does using an LLM help participants make more accurate self-diagnoses and care decisions for common illnesses, compared to their first guess without any help?
• How much better is it when people work together with an LLM, compared to using a regular search engine, using the LLM alone, or how doctors would decide? Researchers will compare participants who were randomly assigned to either the LLM group (using DeepSeek) or the search engine group to see if the LLM-assisted approach leads to better clinical judgments. Participants will:
• Read one of 48 short, realistic health vignettes;
• Make an initial guess about what might be wrong by listing up to three possible causes, ranked from most to least likely, and choose a care level: seek immediate care, see a doctor within one day, see a doctor within one week, or manage at home without medical care.
• Use their assigned tool (either DeepSeek or a standard search engine) to look up information and update their guess and care decision;
• Submit their final diagnosis and care choice after using the tool. In addition, the study team evaluated the performance of four other AI models (GPT-4o, GPT-o1, DeepSeek-v3, and DeepSeek-r1) and 33 experienced general physicians on the same vignettes.

Conditions

Vignette Based Intervention; LLM-based AI Dialogue Bot

Outcome Measures

Primary Outcomes (2)

• Top-3 Diagnostic Accuracy

  The primary diagnostic outcome was defined as the proportion of participants who included the correct diagnosis in their top three differential diagnoses after using the assigned tool (LLM or search engine). Accuracy was assessed for each of the 48 clinical vignettes and aggregated across all participants in each group.

  Immediately after intervention (within the same survey session)

• Triage Accuracy (4-class exact match)

  Triage accuracy was defined as the proportion of participants who selected the correct triage level (emergent care, within one day, within one week, or self-care) that matched the reference standard. There were 12 vignettes per triage category.

  Immediately after intervention (within the same survey session)

Secondary Outcomes (2)

• Top-1 Diagnostic Accuracy

  Immediately after intervention (within the same survey session)

• Triage Accuracy (2-class binary match)

  Immediately after intervention (within the same survey session)

Study Arms (2)

layperson-LLM integrated group

EXPERIMENTAL

After initially answering a clinical diagnosis and triage question without the aid of tools, the participants were asked to use a large language model (Deepseek v3 or r1) to retrieve health information and then answer the same question again

Behavioral: AI-assisted health information seeking

layperson-search engine group

ACTIVE COMPARATOR

After initially answering a clinical diagnosis and triage question without the use of tools, the participants were required to use a search engine to retrieve health information and then answer the same question again

Behavioral: Conventional internet search for health information

Interventions

AI-assisted health information seeking BEHAVIORAL

Participants in this group used a large language model (DeepSeek) to search for medical information related to a clinical vignette after providing initial diagnostic and triage decisions. They were instructed to interact freely with the model to gather insights and then update their diagnoses and triage recommendations. The intervention simulates real-world use of AI tools for personal health decision-making

layperson-LLM integrated group

Conventional internet search for health information BEHAVIORAL

Participants in this group used mainstream internet search engines (e.g., Baidu, Google, Bing) to look up information about the clinical vignette after making initial diagnostic and triage decisions. They were allowed to search freely but were not permitted to use any named AI chatbot or large language model platform. This group represents typical self-directed online health information seeking behavior.

layperson-search engine group

Eligibility Criteria

• Age: 18 Years+
• Sex: all
• Healthy Volunteers: No
• Age Groups: Adult (18-64), Older Adult (65+)

You may qualify if:

• Age 18 years or older
• Current resident of mainland China
• History of high-quality participation in online surveys on Credamo platform (historical survey acceptance rate ≥ 80% and personal credit score ≥ 70)

You may not qualify if:

• Incomplete survey responses
• Failure on embedded quality-check items
• Implausibly short completion time (\<180 seconds for search engine group; \<360 seconds for LLM group)
• Provision of non-diagnostic or irrelevant responses (e.g., "unknown", "don't know")
• Consistent pattern of identical responses across all items

Sponsors & Collaborators

• Huazhong University of Science and Technology: lead

Study Sites (1)

Tongji Medical College of Huazhong University of Science & Technology School of Medicine and Health Management

Wuhan, Hubei, China

Location

Study Officials

• Chenxi Liu

  Huazhong University of Science and Technology

  PRINCIPAL INVESTIGATOR

Study Design

• Study Type: interventional
• Phase: not applicable
• Allocation: RANDOMIZED
• Masking: SINGLE
• Who Masked: PARTICIPANT
• Purpose: HEALTH SERVICES RESEARCH
• Intervention Model: PARALLEL
• Sponsor Type: OTHER
• Responsible Party: PRINCIPAL INVESTIGATOR
• PI Title: Co-Investigator

Study Record Dates

• First Submitted: November 17, 2025
• First Posted: November 26, 2025
• Study Start: April 27, 2025
• Primary Completion: July 1, 2025
• Study Completion: July 1, 2025
• Last Updated: November 26, 2025

Record last verified: 2025-10

Data Sharing

• IPD Sharing: Will not share

Locations

CN (1)