Department of Biostatistics, Epidemiology, and Population Health at the Faculty of Medicine, Public Health and Nursing, Universitas Gadjah Mada, Indonesia
Project Title: Behaviorally Informed Conversational AI to Improve Health Insurance Premium Payment Adherence in Indonesia
Mr. Anis Fuad is a researcher in population health informatics and a faculty member in the Department of Biostatistics, Epidemiology, and Population Health at the Faculty of Medicine, Public Health and Nursing, Universitas Gadjah Mada, Indonesia. His work focuses on the intersection of digital health, artificial intelligence, health financing, and behavior change to strengthen health systems. He has contributed to national and international initiatives on digital health transformation, health information systems, and AI applications for public health. Through his research, he seeks to translate behavioral science into practical digital solutions that improve access to healthcare and support Universal Health Coverage (UHC).
This SABA-funded project builds upon an earlier collaborative study with BPJS Kesehatan, Indonesia’s National Health Insurance Agency, which developed a proof-of-concept generative AI chatbot to support participants with overdue premium payments. Guided by the Transtheoretical Model (TTM), the chatbot tailored conversations according to participants’ readiness to change. The study revealed that individuals at similar stages of readiness often displayed markedly different payment behaviors. While some lacked knowledge about payment procedures or encountered financial and logistical barriers, many expressed strong intentions to pay but repeatedly postponed taking action.
The project will investigate how principles of behavior analysis can strengthen conversational AI by addressing the gap between intention and action. It will examine the effects of antecedent prompts, commitment strategies, immediate feedback, and reinforcement-based messaging within AI-driven conversations to encourage timely premium payments. Particular attention will be given to participants who consistently make commitments but fail to follow through. The study will also identify behavioral patterns that can inform personalized, evidence-based AI conversations tailored to participants' individual circumstances and needs.
The project aims to develop a behaviorally grounded framework for designing conversational AI interventions that support health financing programs. The findings will advance the integration of behavior analysis, AI, and digital public health. They will also provide practical guidance for scaling AI-enabled behavioral interventions to strengthen health insurance participation and support Universal Health Coverage.
Back to International Development Grant Recipients
