The effectiveness of AI-driven health promotion interventions in schools and communities: A systematic review

Artificial intelligence is increasingly being integrated into health education and health promotion, offering opportunities for personalization, behavioural monitoring, and scalable interventions. This systematic review examines the application and effectiveness of AI-driven tools in school and community health education, highlighting their impact, methodological quality, and contextual relevance, especially for LMICs like Nigeria. A comprehensive literature search was conducted following the PRISMA 2020 guidelines across PubMed, Scopus, IEEE Xplore, Google Scholar, and ScienceDirect for studies published between January 2015 and October 2025. Twelve empirical studies met the inclusion criteria, ranging from randomised controlled trials, quasi-experimental, mixed-methods, and pilot designs. Data extraction encompassed intervention type, AI technology, population, outcomes, and key findings. Methodological quality was assessed using the Mixed Methods Appraisal Tool (MMAT, 2018). The result shows that various AI technologies, including chatbots, predictive analytics, intelligent tutoring systems, and wearable sensors, have been employed to improve health literacy in physical activity, nutrition, and disease prevention. Most studies were of high methodological quality (MMAT score 4–5/5) and reported positive behavioural or educational outcomes, including increased health knowledge (↑25–40%), enhanced engagement, and improved preventive health behaviours. At the same time, several challenges persisted concerning digital access, data ethics, and AI literacy among educators. AI-driven interventions hold strong potential for transforming health promotion and education, especially when incorporated within school curricula and community outreach. Ethical governance will have to be strengthened, teacher capacity built, and access equitably provided if implementation is to be sustained in Nigeria and other similar LMICs.