Using Power BI and SQL for Predictive Health Data Analytics to Improve Decision-Making and Performance Tracking in Public Health Systems

This review examines how SQL-based data engineering and Power BI-driven predictive analytics can transform public health decision-making by enabling real-time monitoring, forecasting, and performance optimization across complex health systems. It synthesizes how mature data pipelines, rigorous governance, and advanced feature engineering such as lags, rolling averages, seasonality markers, and spatial aggregation support critical use cases including outbreak early warning, patient flow prediction, stock-out risk identification, and staffing optimization. The paper further evaluates model integration pathways such as AutoML, Python/R, Azure Machine Learning, and in-database scoring, alongside dashboard architectures that incorporate KPI scorecards, drill-throughs, geospatial mapping, and cohort analytics to enhance insight delivery and executive action. Validation considerations including fairness, interpretability, drift detection, and recalibration are explored, as well as implementation strategies involving stakeholder alignment, change management, and analytics maturity frameworks. Finally, the review highlights persistent challenges in data quality, infrastructure, and workforce capacity, and highlights the necessity of privacy-by-design, responsible AI, and equity-centered analytics. Overall, the study provides a concise roadmap for scalable, ethical, and effective deployment of predictive analytics within public health systems.