Abstract
INTRODUCTION: Nearly half of all deaths in children in India aged younger than 5 years are attributable to undernutrition. Reinforcing the caregiver's positive behaviors through multiple channels can lead to improved child nutrition outcomes. We describe the development and piloting of a chatbot to improve nutrition outcomes for children aged 0-12 months. POSHAN DIDI CHATBOT DEVELOPMENT PROCESS: We engaged key stakeholders to understand existing interventions to improve nutrition outcomes and developed a theory of change that included an intervention that provides nutrition-related information directly to beneficiaries. A chatbot, Poshan Didi, was developed to provide individual counseling to mothers with children aged 0-12 months on age-appropriate, nutrition-related topics. The chatbot was piloted in Katni district from February 2019 to October 2019 in 2 phases to investigate the acceptability and feasibility of the chatbot (Phase 1, n=10 mothers) and to assess whether users would continue to be engaged and would engage with both the automated content as well as through free form questions (Phase 2, n=100 mothers). Quantitative and qualitative data were collected in focus group discussions with health care workers and mothers (n=4) and interviews with mothers (n=26). RESULTS: Mothers viewed the chatbot as being a private channel to discuss topics and referred to Poshan Didi's persona as being knowledgeable. Eighty percent of users replied at least once to a chatbot-generated message. Sixty-four percent engaged beyond the standard chatbot content by discussing at least 1 issue with a nurse who responded to escalated messages. CONCLUSION: This work demonstrates how chatbots could create multiple interaction touchpoints between caregivers and health care workers to increase caregivers' access to age-appropriate nutrition counseling and information. The study shows the value of using the responsive feedback approach in the chatbot design and implementation to improve the efficacy of the digital tool.