Conversational artificial intelligence interventions to support smoking cessation: A systematic review and...

Area of Science:

• Digital Health
• Artificial Intelligence in Healthcare
• Public Health Interventions

Background:

• Conversational artificial intelligence (AI), including chatbots and dialogue systems, is an emerging tool for tobacco cessation.
• AI interventions have the potential to provide personalized support and enhance user engagement in cessation programs.

Purpose of the Study:

• To determine the effect of conversational AI interventions on tobacco cessation outcomes in adult smokers.
• To compare AI interventions, with or without standard cessation support, against no intervention, placebo, or active comparators.

Main Methods:

• A comprehensive literature search was conducted across six databases up to June 2022.
• Included studies were randomized controlled trials (RCTs) published since 2005, focusing on sustained tobacco abstinence (≥6 months).
• Random effects meta-analysis was performed on data extracted by two independent authors, including risk of bias assessment.

Main Results:

• Five RCTs with 58,796 participants met the inclusion criteria, utilizing various AI interventions like chatbots in apps and social media, or internet-based avatars.
• Meta-analysis of three studies revealed that participants receiving AI-enhanced interventions were significantly more likely to achieve 6-month smoking abstinence (RR=1.29, 95% CI [1.13, 1.46]).
• High loss to follow-up and overall high risk of bias were noted across the included studies.

Conclusions:

• Limited but promising evidence suggests conversational AI interventions are effective for tobacco cessation.
• While all included studies showed benefits, high heterogeneity necessitates cautious interpretation of results.
• Further well-designed RCTs adhering to standardized reporting guidelines are crucial for this rapidly evolving field.