Public e-governance in healthcare increasingly relies on machine learning (ML) to support policy and clinical decisions. However, the opaque black-box nature of many ML models can undermine trust and accountability. This research investigates the integration of explainable artificial intelligence (XAI) techniques – including SHAP (SHapley Additive exPlanations), LIME (Local Interpretable Model-agnostic Explanations), and others – to enable transparent decision-making in a European public healthcare governance context. We present a case study using public health datasets to predict healthcare outcomes with models like XGBoost for structured data and BERT (Bidirectional Encoder Representations from Transformers) for textual data. A comparative evaluation of explainability frameworks is conducted, assessing their ability to provide both local and global insights. Model performance with XGBoost achieving ~80–85% accuracy and AUC ≈0.83 on the prediction task and illustrate how XAI techniques made model predictions transparent and explaining individual predictions. SHAP and LIME explanations are compared side-by-side, revealing their strengths in interpretability and limitations in stability and scope. The findings demonstrate that explainable ML can enhance stakeholder trust and regulatory compliance in digital public health governance, aligning with European requirements for transparency. Integrating XAI into e-governance systems fosters more accountable, understandable, and thus trustworthy AI-assisted decisions in the public healthcare sector.