Unraveling the drivers of sustainable health system performance through an integrated DEMATEL–PLS-SEM–ANN framework in emerging Indian economies

Abstract Achieving sustainable health system performance (SHP) is a pressing challenge in emerging economies like India, where disparities in infrastructure, technology, workforce, and policy persist. This study investigates the complex, interdependent drivers of SHP by integrating three analytical methods. Decision-Making Trial and Evaluation Laboratory (DEMATEL), Partial least squares structural equation modeling (PLS-SEM), and artificial neural networks (ANN). Guided by socio-technical systems theory and systems thinking, we explore how health infrastructure (HI), technology adoption (TA), health workforce (HW), policy support (PS), and community engagement (CE) influence SHP. Data were collected from 412 stakeholders across six Indian states using a stratified purposive sampling method. DEMATEL identified causal relationships among constructs; PLS-SEM tested hypothesized paths, including mediation by CE and moderation by PS; and ANN validated predictive strength and variable importance. Results reveal that HI, TA, HW, and PS positively affect SHP. CE significantly mediates the impact of HI, TA, and HW, while PS strengthens the effects of TA and HW. TA and PS emerged as the most influential predictors. These findings underscore the need for integrated, community-centered, and policy-supported strategies to strengthen health systems. This study offers a novel multi-method framework to inform evidence-based health reforms in India and similar contexts.