Development and Efficacy Evaluation of an AI Chatbot for Personalized Nutrition Counseling Among Postnatal Mothers: A Mixed-Methods Randomized Controlled Trial

Title: Development and Evaluation of an AI-Powered Multilingual Nutrition Chatbot for Personalized Counseling Among Postnatal Mothers: A Randomized Controlled Trial

Background and Rationale: Inadequate nutrition knowledge and poor dietary adherence are common among postnatal mothers, particularly in underserved communities. Mobile health (mHealth) tools can bridge this gap. This study aims to develop and test an AI-based WhatsApp chatbot that provides personalized, evidence-based nutrition counseling in Kannada, Tamil, Hindi, and English.

Objectives:

1. To assess the effectiveness of a multilingual AI nutrition chatbot in improving maternal nutrition knowledge, dietary diversity, and supplement adherence among postnatal mothers.
2. To evaluate user satisfaction, usability, and engagement with the chatbot.

Study Design: Parallel-group, randomized controlled trial (1:1 allocation) with 150 postnatal mothers recruited from urban and semi-urban Anganwadi centres in Bengaluru.

Intervention: Participants in the intervention arm will receive access to an AI-driven WhatsApp chatbot providing personalized nutrition counseling, reminders, and education messages for 4 weeks. The control group will receive standard government nutrition IEC materials.

Outcomes:

• Primary Outcome: Change in mean nutrition knowledge score from baseline to 4 weeks.
• Secondary Outcomes: Dietary diversity, supplement adherence, breastfeeding self-efficacy, and System Usability Scale (SUS) scores.

Duration: 14 months total (Nov 2025 – Dec 2026); active intervention Jan – Aug 2026.

Sample Size: 150 participants (75 per arm), powered at 80% to detect a moderate effect size (Cohen’s d = 0.5).

Inclusion Criteria: Postnatal mothers (0–12 months postpartum), smartphone users with WhatsApp access, able to read or understand one of the study languages.

Exclusion Criteria: Mothers with severe illness, non-consent, or without mobile access.

Setting: Anganwadi centres and community health facilities under the urban feild practice area

Ethical Approval: Approved by the Institutional Ethics Committee, Sapthagiri Institute of Medical Sciences & Research Centre, Bengaluru (Phone: 080-22188702).

Principal Investigator: Dr. Neha Dsouza, Assistant Professor, Department of Community Medicine, Sapthagiri Institute of Medical Sciences & Research Centre, Bengaluru. Email: drnehadsouza94@gmail.com | Phone: 8618684440