Architecting Synergy: Knowledge Graphs for Representing Complex, Multi-Domain Patient Information

Healthcare systems exchange more data than ever, yet gaps in care persist: missed referrals, unsafe polypharmacy, and loss of continuity. This paper asks: what information architecture supports representing care as a dynamic, multi-domain, intent-driven process? Drawing on empirical failure modes (appointment scheduling, referrals, labs, medication processes) identified in primary care in Canada, along with other care failures (e.g. guideline misapplication, discharge lapses), we derive functional and technical requirements. We evaluate four candidate architectures against these requirements. We find that only knowledge graphs satisfy the combination of semantic linkage, schema flexibility, state tracking, longitudinal reasoning, and explainability. We then discuss regulatory, governance, and vendor implications. Knowledge graphs offer a promising path forward toward truly interoperable, safe, and patient-centered care.