PRIVACY-AWARE REQUIREMENTS TRACEABILITY AND RISK INTELLIGENCE FOR INTEROPERABLE DIGITAL HEALTH INFORMATION SYSTEMS

Interoperable digital health systems connect electronic health records, patient portals, clinical services, administrative platforms, and third-party applications, creating dependencies that can propagate requirement changes across workflows, data exchanges, privacy constraints, and security controls. This study develops a privacy-aware requirements traceability and risk intelligence framework to identify such dependencies and assess associated risks. The framework represents requirements, components, workflows, data elements, privacy constraints, security controls, and organizational responsibilities in a typed multi-relational graph. Analytical risk scoring is combined with explainable machine learning to detect high-risk states and identify contributing factors. Evaluation uses 10 synthetic healthcare architectures, 15 public standards and implementation documents, and 120 simulated requirement change scenarios. The proposed graph detected conflicts in 112 of 120 simulated requirement change scenarios, corresponding to a 93.3% detection rate. The combined analytical and ML model achieved 86.3% precision, 84.1% recall, and 85.2% F1 score. These findings show that integrated traceability can support change impact analysis, privacy assessment, and interpretable risk review. Future work should test the framework with real healthcare systems and organizational incident data.