Unguided Chatbot-Delivered Cognitive Behavioural Intervention for Problem Gamblers Through Messaging App: A...

Area of Science:

• Psychology
• Digital Health
• Behavioral Science

Background:

• Problem gambling affects a significant portion of the population.
• Many problem gamblers are hesitant to seek traditional face-to-face support.
• Internet-delivered interventions offer a potential alternative, but often suffer from high dropout rates.

Purpose of the Study:

• To develop and evaluate GAMBOT, a low-dropout, unguided intervention integrated into a messaging app for problem gamblers.
• To assess the efficacy of GAMBOT in reducing gambling severity and symptoms.

Main Methods:

• A randomized, quadruple-blind, controlled trial was conducted.
• Participants received either the GAMBOT intervention (daily monitoring, personalized feedback, cognitive behavioral theory-based messages via app) or assessment-only messages for 28 days.
• Primary outcome: change in Problem Gambling Severity Index (PGSI); Secondary outcome: change in Gambling Symptom Assessment Scale (G-SAS).

Main Results:

• No significant difference was found between groups for the primary outcome (PGSI).
• A significant reduction was observed in the secondary outcome (G-SAS) for the intervention group.
• The GAMBOT intervention demonstrated a very low dropout rate (6.7%) with high participant engagement (77% completion).

Conclusions:

• Integrating interventions into familiar messaging apps via chatbots can help overcome high dropout rates in unguided digital health solutions.
• GAMBOT shows potential for engaging problem gamblers who avoid traditional support, particularly in reducing gambling-related symptoms.
• Further research is needed to enhance chatbot content for greater engagement and effectiveness in addressing problem gambling.