Joining up health and social care: what we learned prototyping the Social Care Interoperability Platform

Working with DHSC and NHS England, Synanetics helped explore how information can move more effectively between adult social care and health systems. Through user research, prototyping and supplier testing, the SCIP programme demonstrated how standards-based interoperability could help information follow the person, improving continuity of care.

👉 Read more...

Posts

Traditional data modelling organises information into predefined structures. AI retrieval changes that assumption. This article explores why techniques such as chunking, embeddings and vector search are becoming equally important design considerations, and how healthcare organisations can combine them with standards like FHIR to create more effective AI-powered experiences.
AI agents are often presented as autonomous assistants, but real-world healthcare workflows rarely fit neatly into a single agent model. This article explores why effective AI solutions depend on collaboration between specialised agents, orchestration services and governed workflows, and what this means for the design of healthcare applications.
In a previous post, I described how we created a semantic layer over FHIR to support AI-driven capabilities such as semantic search and risk stratification. The response prompted a number of follow-up questions, many of them less about AI itself and more about the architectural decisions behind it: Why use FHIR narratives? Why are embeddings useful? What role does retrieval play? Why does chunking matter? And what exactly changes when agents are introduced into the architecture?
There is no shortage of discussion about the opportunities that AI may create within health and social care. Modern language models appear increasingly capable of interpreting narrative, answering questions and identifying relationships across both structured and unstructured information. In a domain where information is often recorded and understood as a mixture of records, observations and stories, it is tempting to conclude that the way data is represented no longer matters. In other words, perhaps structured data is dead.
The Social Care Interoperability Platform (SCIP) Hack Day brought together six digital social care record providers to test the practical realities of interoperability using a prototype federated integration platform deployed specifically to support the event. Working from a shared data model (MODS) and transforming to and from FHIR, teams were able to publish, retrieve, and consume synthetic data end-to-end within a single day.
Recent work testing federated data sharing in an adult social care context has demonstrated how established NHS interoperability principles can be extended beyond traditional healthcare settings. Using a derivative of an existing NHS-owned platform and working collaboratively with digital social care record suppliers, this approach has shown how data can be discovered, accessed, and combined across organisational boundaries in practice. While joining up data across health and care remains challenging due to fragmentation and differing ownership, data federation offers a scalable alternative to centralised repositories and point‑to‑point integrations, enabling data to remain with providers while being accessed through a unified interface as if it were a single dataset.