Can Mobile Health Tools Increase Safe Health-Seeking Behaviour? A Mixed-Methods Exploration

High maternal mortality rates persist as a critical challenge in low-resource settings. This study investigates whether integrating a Clinical AI (cAI) designed for symptom assessment (Ada) into a Whatsapp-based government platform (MomConnect) enhances self-care and health-seeking decision-making among pregnant women and mothers. From May to June 2023, a mixed-methods study enrolled participants. A baseline survey captured initial care-seeking intentions, followed by cAI use and a one-week follow-up survey on actual behaviors. A physician panel reviewed randomly selected cases, comparing “most appropriate” and “safe” urgency ratings with the cAI’s advice, as well as participants’ intentions and actions. Among 968 participants, 44.8% changed their initial care-seeking intentions post-cAI, increasing safe decisions by 32.1% and appropriate decisions by 41.3%. The physician panel deemed the cAI’s advice safe in 98.4% of reviewed cases. A total of 86.3% faced multiple healthcare barriers, correlating with higher unsafe intentions. However, post-cAI, their actual behaviors matched those with fewer or no barriers. Integrating a cAI into a government platform positively influenced maternal decision-making, reducing disparities in healthcare access. These findings underscore the value of AI-driven digital health tools in delivering timely, equitable care.