Pilot Study of an AI Chatbot for Reducing Alcohol Use in Adults with Metabolic Alcohol-Related Liver Disease

What this Trial Is About

Researchers are evaluating the feasibility and acceptance of a large language model LLM-based chatbot aimed at helping patients with metabolic alcohol-related liver disease MetALD reduce their alcohol consumption. This pilot randomized controlled trial aims to assess participant enrollment, retention rates at three and six months, user compliance, usability of the chatbot, and its effect on reducing alcohol use. The study also explores participants experiences and perceptions of interacting with the chatbot. Participants will be randomly assigned to one of two groups for a three-month period following randomization. One group will interact with the LLM-based chatbot, which delivers four personalized sessions about alcohol reduction and is accessible anytime for additional support. The comparison group will receive four individual telephone counseling sessions from trained counselors offering personalized advice on reducing alcohol intake. Both groups will also receive a brief alcohol intervention consisting of short counseling sessions and a self-help booklet. During the study, researchers will monitor recruitment and retention rates, weekly alcohol consumption, and patterns of binge and heavy drinking through assessments at three and six months after randomization. Additional measures include alcohol use disorder identification test AUDIT scores and participants plans for drinking in the upcoming month. Participants are expected to engage in behavioral assessments and provide feedback on treatment acceptability. The study is designed to last approximately six months per participant, with continuous monitoring of alcohol use and treatment adherence.

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