CClinicalTrials.gg
Not yet recruitingNCT07844044LAMP-2Updated Sep 28, 2026

3M-6D Educational Intervention to Improve LLM-Assisted Medical Consultation in Neurology Outpatients

An interventional study of three minutes six dimensions education and ChatGPT in Condition Category Concordance, sponsored by Capital Medical University. Not yet recruiting at 3 sites in China. Open to participants aged 18 Years and older. Per ClinicalTrials.gov, last updated 2026-09-28.

Sponsored by Capital Medical University · Not applicable, Interventional, and Health services research

Phase
Not applicable
Study type
Interventional
Enrollment
396
Allocation
Randomized
Ages
18 Years and older
Sex
All
01

Study summary

This study will evaluate whether three-minute six-dimension education (3M-6D education) can improve the reliability of large language models (LLMs) as medical assistants in neurology outpatients. Participants will be randomly assigned to receive or not receive 3M-6D education before using ChatGPT for medical consultation.

Read the detailed description

This randomized, controlled clinical trial will evaluate whether three-minute six-dimension education (3M-6D education) can improve the reliability of large language models (LLMs) as medical assistants in neurology outpatients.

Eligible participants will be randomly assigned in a 1:1 ratio to the 3M-6D education GPT group or the GPT group. Participants in the 3M-6D education GPT group will first receive approximately three minutes of structured education and then use ChatGPT for medical consultation. Participants in the GPT group will directly use ChatGPT for medical consultation without receiving 3M-6D education. All participants will communicate with ChatGPT using unrestricted natural language.

The 3M-6D education is a structured framework designed to help patients describe their symptoms and provide more complete medical information during LLM-assisted medical consultation. It guides patients to describe six key dimensions.

02

Conditions studied

  • Condition Category Concordance

Keywords

  • 3M-6D education
  • Large Language Models
  • Cognitive Load Theory
  • neurology outpatients
  • condition category concordance
03

Who can participate

Ages eligible
18 Years and older
Sexes eligible
All
Accepts healthy volunteers
No

Inclusion criteria

  1. Aged ≥ 18 years
  2. Completed primary school or higher education;
  3. Able to use a smartphone or computer to complete online interactions;
  4. Presenting to the hospital for neurological symptoms;
  5. This visit is the first consultation for the current symptoms;
  6. Has not used AI tools or online resources to assess the current condition prior to this visit;
  7. Able to understand and comply with study procedures and to provide written informed consent.

Exclusion criteria

Exclusion Criteria:

  1. Currently or previously employed as a healthcare worker;
  2. Previously received systematic medical training;
  3. Condition requires emergency care;
  4. Participating in another clinical trial that may affect this study's outcomes;
  5. Any other condition the investigators consider unsuitable for enrollment or likely to interfere with trial results.
04

Study design

Phase
Not applicable
Primary purpose
Health services research
Allocation
Randomized
Intervention model
Parallel assignment
Masking
Single (Outcomes assessor)
Enrollment
396 participants (estimated)

Study arms

  • Experimental
    3M-6D education GPT Group

    Participants will first be trained in 3M-6D education, then use ChatGPT to complete a consultation task in unrestricted natural language.

    Behavioral: three minutes six dimensions education · Other: ChatGPT

  • Active comparator
    GPT Group

    Participants will use ChatGPT to complete a consultation task in unrestricted natural language.

    Other: ChatGPT

Interventions

  • Behavioralthree minutes six dimensions education

    3M-6D education is designed based on Cognitive Load Theory to reduce the cognitive burden on patients during medical interactions with AI and to improve the clarity and completeness of symptom reporting. Guided by cognitive load theory and the natural process physicians use to take medical histories, the investigators identified candidate information dimensions and developed a structured expression framework with six dimensions for public health queries through a Delphi expert consensus process. Participants were instructed to use the framework to describe their symptoms across these six dimensions; this process can typically be completed within three minutes, so the investigators call this approach three minutes six dimensions education (3M-6D education).

    Also known as: 3M-6D education

  • OtherChatGPT

    Participants use ChatGPT to complete a consultation task in unrestricted natural language.

05

What researchers measure

Primary outcomes

  1. Condition category concordance

    The primary outcome of the LAMP-2 trial was condition category concordance, defined as at least one condition listed in a participant's response falling within the same prespecified neurological disease category as the final reference diagnosis.

    Time frame: 30 days

Secondary outcomes

  1. Preparation for Decision Making Scale

    Preparation for Decision Making Scale (PrepDM) is a self-reported scale used to assess participants' readiness for medical decision-making after medical consultation. Higher scores indicate greater preparation for decision-making.

    Time frame: Within 1 hour

  2. Consultation duration

    Consultation duration is defined as the time from the patient's arrival at the outpatient clinic to the completion of the outpatient visit.

    Time frame: Within 1 hour

  3. Patient satisfaction with this consultation

    Patient satisfaction is a self-reported measure used to assess participants' satisfaction with the outpatient visit. Higher scores indicate greater satisfaction.

    Time frame: Within 1 hour after the outpatient visit

  4. Red-flag identification

    Red-flag identification is defined as the proportion of participants whose final response includes the key warning signs predefined by experts for the participant's presenting symptoms.

    Time frame: Within 1 hour after the outpatient visit

  5. NASA-TLX score

    NASA-TLX score is a self-reported task-load score measured immediately after the LLM-assisted medical consultation with ChatGPT. It includes six domains: mental demand, physical demand, temporal demand, effort, frustration, and performance. Each domain is scored from 0 to 100. The total score is the mean of the six domains. Higher scores indicate greater perceived task load.

    Time frame: Within 1 hour

06

Study locations

3 sites
  • Aerospace Center Hospital
    Beijing, Beijing Municipality, China
  • Huimin Hospital
    Beijing, Beijing Municipality, China
  • Mentougou District Hospital
    Beijing, Beijing Municipality, China
07

References and documents

Individual participant data

Plan to share: Undecided

No publications or documents are linked to this record.

08

Registry details

Key details

Study ID
NCT07844044
Lead sponsor
Capital Medical University
Responsible party
Ji Xunming,MD,PhD (Principal Investigator, Capital Medical University) — Principal investigator
First posted
Sep 28, 2026
Start date
Sep 26, 2026 (estimated)
Primary completion
Oct 26, 2026 (estimated)
Completion
Oct 26, 2026 (estimated)
Last update
Sep 28, 2026

Study contacts

Xunming Ji
Contact
jixm@ccmu.edu.cn
01083198962
Chuanjie Wu
Contact
wuchuanjie@ccmu.edu.cn
01083199439

Oversight

Data monitoring committee
Yes
FDA-regulated drug
No
FDA-regulated device
No
View the source record on ClinicalTrials.gov ↗

Not currently enrolling

This study is not yet recruiting, as verified in Sep 2026. You cannot join it, but the record below documents what was studied.

Follow this study

Get an email when the registry record changes — status, dates, results — or when someone posts here.

Sign in to follow

Discussion

Questions and observations about this study, from anyone following it. Not medical advice, and not a channel to the study team — their contact details are on the registry record.

Sign in to join the discussion. Reading takes no account; posting does. You choose a display name, and a pseudonym is the default.

Nothing here yet. If you are running this trial, taking part in it, or weighing whether to, this is the place to say so.

Start the discussion