The digital health enterprise: one architecture for the building and the clinic
PUBLIC SECTOR · HEALTH · TECHNOLOGY
The situation
A government health service was about to commit to a decade of technology and capital investment, and had no agreed picture of what it was building towards. Decisions were being made system by system, each defensible on its own terms and none of them adding up. Clinical systems, diagnostic systems, corporate systems and the building services in the hospitals themselves were being planned by different people, procured on different cycles, and connected to each other by whatever the vendor of the day happened to support. The organisation had a clear clinical vision and a set of service principles it genuinely believed in, and no way of testing whether a technology decision advanced them or cut across them.
The immediate task was to produce an end state framework that could guide three months of detailed strategy work. The real task was to give the organisation a way of making technology decisions that a clinician, a chief financial officer and an engineer could all recognise as legitimate.
Before
Decisions made system by system
Buildings and clinical systems planned apart
Proprietary building networks
Identity managed twice, physical and logical
Benefits asserted rather than measured
The work
Retained with a group of technology partners to write the framework. It answered four questions in sequence, and the sequence is the point:
How does this contribute to what the organisation is already trying to do. The technology vision was derived from the health service’s own vision rather than written alongside it, and the scope of technology was defined deliberately broadly as any device or system that exchanges data. That single definition is what later allowed building services to be pulled inside the architecture rather than left outside it.
What does it mean for each stakeholder. Benefits were set out separately for patients, clinicians, support functions and management, with the standing requirement that every benefit had to be justified by the plan and be measurable. Vague benefit language is how technology programmes of this size escape accountability, and closing that door at the framework stage is much easier than closing it later.
What rules will govern decisions. Ten principles for technology decision making, each with a statement and a rationale: patient centred, safe, accessible, private and confidential, integrated, compliant, reliable, sustainable, responsive and flexible, and affordable. Each was mapped explicitly onto the health service’s existing clinical service principles, so that technology decisions inherited clinical priorities instead of competing with them. Each was also to carry an implications statement setting out the tasks, resources and potential costs of actually following it.
What does it take to be successful. Nine integrated building blocks defined down to component level: clinical and point of care, diagnostics, research and education, community and cross jurisdiction connectivity, clinical and process support, corporate functions, building management, access devices, identity and access management, unified communications, enterprise integration, a medical grade network, and the data centre.
The consequential decision was to bring building management inside the architecture rather than treat it as facilities engineering. That meant requiring building subsystems to sit directly on internet protocol infrastructure rather than proprietary networks, to support open standards for interoperability, to present a common user interface so operators could be cross trained, and to be operated centrally or virtually rather than from separate control rooms. It also meant a single identity across the physical and logical estate, location services covering staff, patients and critical assets, and condition based rather than calendar based maintenance driven by a real time view of the facility.
After
Ten principles governing every decision
One converged enterprise architecture
Internet protocol and open standards throughout
Single identity across the whole estate
Benefits defined per stakeholder, each measurable
The outcome
An agreed end state picture that gave three months of detailed strategy work something to aim at, and gave every subsequent procurement a test to be measured against.
Ten technology decision principles, each traced back to a principle the clinical organisation had already adopted, which is what made them survive contact with clinicians rather than being dismissed as a technology document.
Nine building blocks defined to component level, integrated with each other and aligned to models of care, building infrastructure and workforce, so that the architecture described an operating environment rather than a systems inventory.
And a discipline built into the framework, benefits isolated and measurable. Principles had to carry a statement of what following it would cost in tasks, resources and money.