Real-time Smoking Cessation Instant Messaging Support Using a Large Language Model (LLM)-Based Chatbot Via "Quit to Win" 2025 (QTW2025)

Summary

The goal of this trial is to learn if chatbot-based instant messaging works to help smoking cessation in general adult smokers. It will also learn about the experience, attitude, and perception of using an LLM-based chatbot. The main questions it aims to answer are:

1. Will LLM-based chatbot smoking cessation intervention have a higher validated abstinence rate than the control group?
2. Will LLM-based chatbot smoking cessation intervention have a higher self-reported abstinence rate, smoking reduction rate, and smoking cessation services use rate than the control group?

Researchers will compare LLM-based chatbot smoking cessation intervention to a usual care group (brief advice based on AWARD and personalized active referral) to see if chatbot-based instant messaging support works to promote smoking cessation.

Participants in the intervention group will receive:

1. AWARD advice
2. Personalized active referral
3. 12 weeks of chatbot-based instant messaging support (via WhatsApp)

Conditions

• Smoking Cessation

Interventions

BEHAVIORAL

Referral card

The contents consist of brief information and a highlight of existing smoking cessation services, contact methods, motivation information and strong supporting messages or slogans.

BEHAVIORAL

AWARD advice

AWARD advice include Ask about smoking history, Warn about the high risk, Advise to quit, Refer smokers to smoking cessation services (with a referral card), and Do it again.

BEHAVIORAL

Brief leaflet on health warning and smoking cessation

The contents of the leaflet include (1) highlights of the absolute risk of death due to smoking; (2) the whole list of diseases caused by active and second-hand smoking; (3) ten horrible pictorial warnings of health consequences of smoking and second-hand smoking in one page to maximize the impacts; (4) benefits of SC and (5) simple messages to encourage participants to quit smoking.

BEHAVIORAL

Self-help smoking cessation booklet

The contents include information about the benefits of quitting, smoking and diseases, methods to quit, how to handle withdrawal symptoms, declaration of quitting, etc.

BEHAVIORAL

Personalized active referral

Smokers will be introduced to various SC services in Hong Kong (using the referral card) and be motivated to use these services. Well-trained SC ambassadors will assist smoker to choose favourite/most convenient or preferred type of services. Research staff will assist participants in booking or re-booking the SC services at the 1- and 2-month follow-ups (after very brief questionnaire surveys). Participants' contact information will be forwarded to SC services providers within seven days, and providers are expected to contact the participants within 1-2 weeks. Research staff will also monitor SC services use of the participants at each follow-up (1-, 2-, 3- and 6- month) and assist participants to book or re-book the appointments if necessary at 1- and 2- month follow-up. We shall liaise with the existing service providers and seek their assistance in helping our smokers in helping our smokers in a timely manner.

BEHAVIORAL

12 weeks of chatbot-based instant messaging support

Participants in the intervention group will receive 12 weeks of instant messaging support delivered by the LLM-based chatbot accessible via WhatsApp platform. The chatbot, powered by GPT-4o model (or more advanced model available) using prompt-engineering and agent-based techniques, will deliver brief theory-based, structured SC intervention alongside freeform, on-demand support. The structured intervention session deploys the 5As model (Ask, Advise, Assess, Assist, and Arrange follow-up) and 5Rs model (Relevance, Risks, Rewards, Roadblocks, Repetition), as used in our previous telephone-counselling trials and recommended by WHO for brief SC intervention.

BEHAVIORAL

Reminder messages

WhatsApp messages on follow-up survey reminders.

Sponsors & Collaborators

• The University of Hong Kong

  lead OTHER

Principal Investigators

• Man Ping Wang, PhD · The University of Hong Kong

Study Design

Allocation

RANDOMIZED

Purpose

HEALTH_SERVICES_RESEARCH

Masking

SINGLE

Model

PARALLEL

Eligibility

Min Age

18 Years

Sex

ALL

Healthy Volunteers

No

Timeline & Regulatory

Start

2025-06-21

Primary Completion

2026-06-30

Completion

2027-06-30

Countries

• Hong Kong

Study Locations