Related Experiment Video
Updated: Oct 9, 2026
Virtual Agent for Real-Time Motivational Interviewing by Integrating Adaptive Nonverbal Behavior and Language Models
Published on: December 23, 2025
Codevelopment, Implementation, and Evaluation of an AI-powered Conversational Interface for Tobacco Cessation:
Monika Arora1, Rajmohan Panda1, Shalini Bassi1
1Public Health Foundation of India, New Delhi, India.
Background:
Despite several efforts to expand tobacco cessation services through tobacco cessation centers (TCCs) and quit lines, key operational challenges still persist. Recently, AI-based digital interventions have shown promise globally for smoking cessation; however, they remain underused in tobacco cessation strategy in the Indian context.
Objective:
This study aimed to (1) codevelop the conversational interface with beneficiaries and cessation providers; (2) assess its feasibility, acceptability, usability, and user engagement; and (3) evaluate its effectiveness in promoting attempts and intention to quit tobacco.
Methods:
The study consists of three phases: (1) codeveloping the comprehensive AI-powered Conversational Interface to Quit Tobacco (CARE) conversational interface through in-depth interviews with tobacco users, counselors, and health care professionals; (2) feasibility testing with tobacco users to assess engagement, usability, and acceptability; and (3) 6-month effectiveness testing using pre-post surveys. The CARE conversational interface will be developed using a retrieval-augmented generation-based large language model by the Indian Institute of Technology (Bombay, Maharashtra, India), delivering personalized, multilingual cessation support via a chatbot integrated into a mobile app. Key evaluation measures will use validated tools such as the Fagerström Test for Nicotine Dependence, Smoking Self-Efficacy Questionnaire, Decisional Balance Scale, and a Knowledge Score Questionnaire. Data will be analyzed using mixed methods, including thematic analysis for qualitative data and descriptive statistics and a multivariate logistic regression for quantitative data.
Results:
The CARE study was funded in March 2025 and is being implemented across 3 TCCs. Preparatory activities, including tool development, site engagement, ethics approval, and trial registration (CTRI/2024/11/076916), were completed in 2025. Phase 1 (codevelopment) is scheduled from February to August 2026, followed by phase 2 (feasibility testing) from September 2026 to August 2027, and phase 3 (effectiveness assessment) from September to November 2027. Data analyses are expected to be completed by late 2027, with key findings on feasibility, acceptability, usability, user engagement, and preliminary cessation outcomes targeted for peer-reviewed publication from 2027 onward.
Conclusions:
The CARE study will attempt to introduce a novel, culturally tailored conversational interface targeting both smokers and smokeless tobacco users in India by integrating AI-based solutions as an adjunct to conventional counseling at the TCCs. Given the single-group pre-post design without a control arm, feasibility and effectiveness findings will be interpreted as preliminary and hypothesis-generating rather than causal evidence of intervention impact. The intervention approach of codevelopment, strengthening capacity in evidence-based content, and enabling a multilingual conversation interface is expected to enhance tobacco user engagement and improve cessation outcomes. Findings from this pilot will inform a future randomized controlled trial and provide evidence for the potential integration of the CARE interface into India's tobacco cessation platforms, offering a low-cost, high-impact solution for India and other low- and middle-income countries that face similar challenges.
Trial Registration:
Clinical Trials Registry-India CTRI/2024/11/076916; https://tinyurl.com/3vfs5d8s.
International Registered Report Identifier (Irrid):
PRR1-10.2196/82264.
