Readiness to Adopt Artificial Intelligence and Big Data Analytics in Primary Health Care for Chronic Disease Management: A Narrative Review with Implications for Saudi Arabia

Abstract

Non-communicable diseases account for most deaths in Saudi Arabia, and primary health care (PHC) centres carry the main responsibility for their detection and control. Artificial intelligence (AI) and big data analytics (BDA) are promoted as tools for risk stratification, screening and population health management, but their value depends on whether health systems are ready to adopt them. To synthesise evidence on the value of AI and BDA for chronic disease management in primary care and on the individual, organisational, technological and environmental factors that shape readiness to adopt them, with attention to Saudi Arabia and Al Madinah. A narrative review of peer-reviewed literature and official policy documents, focusing on 2019-2026, was conducted using the Technology Acceptance Model and the Technology-Organisation-Environment framework as the analytic lens. Bibliographic details and key findings of all cited sources were checked against publisher or official records. The strongest evidence of benefit concerns AI-supported diabetic retinopathy screening, including a randomised trial showing higher screening completion. For risk prediction, apparent advantages of machine learning over conventional models largely disappear in low-bias comparisons, external validation is rare, and few randomised trials exist, none reported from the Gulf. Professionals’ attitudes are positive but conditional; Saudi studies report high trust or perceived usefulness alongside limited knowledge, high fear of replacement and low use (around 26-32%). Organisational factors, including leadership, culture, training and financing, and technological factors, notably incomplete EHR coverage, data quality and interoperability, emerge as the main constraints. Saudi Arabia has established national AI strategy, ethics principles, data protection law and device guidance, but evidence on how these reach PHC is limited, and no study of AI or BDA readiness among practising PHC professionals in Al Madinah was identified. Readiness, rather than algorithmic capability, is the binding constraint on AI and BDA adoption in PHC. Context-specific, theory-based readiness assessment combining individual and organisational dimensions is needed to guide safe and sustainable adoption in Saudi health clusters.