A Large Language Model-based Chatbot for Alcohol Reduction in Patients With Metabolic Alcohol-Related Liver Disease

A Large Language Model-based Chatbot for Alcohol Reduction Counseling in MetALD Patients: A Pilot Randomized Controlled Trial

The goal of this pilot randomized controlled trial is to evaluate the trial feasibility and acceptability of LLM-based chatbot for reducing alcohol use among patients with metabolic alcohol-related liver disease. Specific objectives include:

1. To assess how many MetALD patients accept the invitation to participate in the trial
2. To assess the retention rate of the participants through 3 and 6 months after treatment initiation
3. To assess the acceptability of the LLM-based chatbot in terms of participants' compliance and usability rating
4. To estimate the intervention effect on alcohol reduction
5. To explore the participants' perception and experiences in the chatbot

Study Overview

• Status: Recruiting
• Conditions: Metabolic Dysfunction-Associated Steatotic Liver Disease; Metabolic Alcohol-Related Liver Disease
• Intervention / Treatment: Behavioral: LLM-based chatbot; Behavioral: Alcohol Brief Intervention; Behavioral: Telephone counseling

• Study Type: Interventional
• Enrollment (Estimated): 50
• Phase: Not Applicable

Contacts and Locations

Study Locations

• Hong Kong
  • Hong Kong, Hong Kong
    • Recruiting
    • Hong Kong Queen Mary Hospital
    • Contact: Man Ping Wang, Prof; Phone Number: +85239176636; Email: mpwang@hku.hk

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult; Older Adult
• Accepts Healthy Volunteers: No

Description

Inclusion Criteria:

1. Hong Kong residents aged 18 years or above 2. Diagnosed of MASLD:

1. Evidence of hepatic steatosis (fat accumulation in the liver) confirmed by imaging techniques such as ultrasound, computed tomography (CT), magnetic resonance imaging (MRI), or liver biopsy

2. Presence of metabolic dysfunction, indicated by at least one of the following:

   • Overweight or obesity (body mass index [BMI] ≥25 kg/m²),

     • Type 2 diabetes mellitus, ③Dyslipidemia (elevated triglycerides [≥150 mg/dL] or low high-density lipoprotein [HDL] cholesterol [<40 mg/dL for men, <50 mg/dL for women]), ④Hypertension (blood pressure ≥130/85 mmHg or use of antihypertensive medication).

       • Metabolic syndrome as defined by established criteria (e.g., International Diabetes Federation [IDF], Adult Treatment Panel III [ATP III]) 3. Alcohol consumption: males who drink 210-420 g/week (≈263-525 ml), female who drink 140-350 g/week (≈175-438 ml) 4. Intention to reduce/quit alcohol 5. Able to read and communicate in Chinese 6. Own a smartphone with internet access

Exclusion Criteria:

1. Diagnosed with mental disease or cognitive impairments, or
2. Participating in other ongoing clinic trials

Study Plan

How is the study designed?

Design Details

• Primary Purpose: Treatment
• Allocation: Randomized
• Interventional Model: Parallel Assignment
• Masking: Single

Number of Arms

2

Arms and Interventions

Participant Group / Arm	Intervention / Treatment
Experimental: LLM-based chatbot; LLM-based chatbot + Alcohol Brief Intervention	Behavioral: LLM-based chatbot; Over the three months following randomisation, an LLM-based chatbot will interact with each participant individually to deliver four personalized sessions of alcohol reduction information and advice, and participants will also be allowed to contact the chatbot at any time.; Behavioral: Alcohol Brief Intervention; Brief counselling (5 to 10 mins) to reduce drinking and a self-help booklet
Active Comparator: Telephone counseling; Telephone counseling + Alcohol Brief Intervention	Behavioral: Alcohol Brief Intervention; Brief counselling (5 to 10 mins) to reduce drinking and a self-help booklet; Behavioral: Telephone counseling; Over the three months following randomisation, a trained counsellor will conduct four individual daytime telephone sessions with each participant to provide personalised information and advice on reducing alcohol consumption.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Retention rate	Number of participants completed the follow-up divided by the number of participants	6 months after randomisation
Weekly alcohol consumption	Assessed by questions 1 and 2 of AUDIT-C	3 months after randomisation
Recruitment rate	Number of participants divided by the number of eligible subjects	Through recruitment completion, about 12 months

Secondary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Weekly alcohol consumption	Assessed by questions 1 and 2 of AUDIT-C	6 months after randomisation
AUDIT score and level	Assessed by the Alcohol Use Disorders Identification Test	3 months after randomisation
AUDIT score and level	Assessed by the Alcohol Use Disorders Identification Test	6 months after randomisation
Eligibility rate	Number of eligible patients divided by the number of patients screened	Through recruitment completion, about 12 months
Frequency of binge drinking in the past 30 days	Self-reported	3 months after randomisation
Frequency of binge drinking in the past 30 days	Self-reported	6 months after randomisation
Frequency of heavy drinking in the past 30 days	Self-reported	3 months after randomisation
Frequency of heavy drinking in the past 30 days	Self-reported	6 months after randomisation
Planned drinking in the coming 30 days	Self-reported	3 months after randomisation
Planned drinking in the coming 30 days	Self-reported	6 months after randomisation

Collaborators and Investigators

• Sponsor: The University of Hong Kong

Study record dates

Study Major Dates

• Study Start (Estimated): March 18, 2026
• Primary Completion (Estimated): May 22, 2027
• Study Completion (Estimated): July 22, 2027

Study Registration Dates

• First Submitted: February 26, 2026
• First Submitted That Met QC Criteria: February 26, 2026
• First Posted (Actual): March 4, 2026

Study Record Updates

• Last Update Posted (Actual): March 23, 2026
• Last Update Submitted That Met QC Criteria: March 18, 2026
• Last Verified: March 1, 2026

More Information

Terms related to this study

• Keywords: Chinese; drinking; mHealth; cirrhosis; WhatsApp; Large language model; COM-B model; BCTs; AI-based chatbot
• Additional Relevant MeSH Terms: Pathologic Processes; Pathological Conditions, Signs and Symptoms; Fibrosis; Organic Chemicals; Alcohols; Ethanol
• Other Study ID Numbers: MetALD

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: UNDECIDED

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No