STATISTICAL CAUSAL INFERENCE WITH MACHINE LEARNING FOR GLOBAL HEALTH POLICY OPTIMIZATION

We examine how machine learning driven policy analytics improves global health policy optimization through predictive epidemiological modeling, causal impact estimation algorithms, and integrated health data systems. We develop and test the Global Health Policy Optimization Model using the International Health Policy Impact Dataset covering institutional health policy analytics indicators between 2020 and 2025 in Ghana. The empirical design combines institutional data from public health agencies, hospitals, policy units, international health organizations, and digital health data science units with an expert sample of fifty professionals involved in epidemiological modeling and policy evaluation. Results show that predictive modeling, causal estimation algorithms, and integrated health data systems significantly improve policy effectiveness, resource allocation efficiency, public health risk reduction, and health system resilience. Institutional governance capacity strengthens these relationships by enabling analytical evidence to translate into operational policy decisions. The findings introduce an integrated analytical architecture linking machine learning analytics and governance capacity to policy performance. The framework provides practical guidance for governments seeking to strengthen evidence driven health policy systems and supports global debates on data driven public health governance.