Integrating artificial intelligence (AI) into maternal and neonatal health (MNH) offers significant opportunities for enhancing patient care through advanced predictive modeling, early disease diagnosis, and ongoing monitoring of conditions such as preeclampsia or gestational diabetes. However, significant challenges in economic valuation persist, including data scarcity, complexity, and the nascent stage of AI implementation in clinical practice. There has been no consolidated empirical proof directly justifying widescale AI application in MNH so far, despite its potentially significant economic benefits and direct cost savings. This review demonstrates that AI systems can mitigate adverse drug reactions (ADRs) and enhance the operational efficiency of organizations. As the full economic potential has yet to be understood and quantified, this review examines several existing economic evaluation frameworks: Cost-Effectiveness Analysis (CEA), Cost-Utility Analysis (CUA), Cost-Benefit Analysis (CBA), and Budget Impact Analysis (BIA). A crucial gap exists between rapid technological advancements and robust economic evaluations, further compounded by a lack of standardized reporting frameworks that hinder the synthesis of available evidence. In addition, the review addresses key challenges, including how they affect the healthcare workforce and the economic impact of systemic errors and security breaches, and then discusses the clinical and liability risks posed by “black box” models. Furthermore, the frequent updates essential for the clinical efficacy and safety of AI tools in MNH are often tied to subscription-based models, creating significant financial strain, particularly in low and middle-income-countries (LMICs). To bridge this crucial research gap and the absence of uniform reporting, this paper proposes the AI-MNH economic evaluation lifecycle and a tailored CHEERS checklist. This multi-phase framework is designed to guide comprehensive, long-term economic evaluations and the adoption of a consolidated, standardized approach to support evidence-based policymaking and sustainable resource allocation.
Mohamed A. Ismail (Sat,) studied this question.