A Sri Lankan national health-data blueprint

The programme

Sri Lanka, a country of roughly 22 million with a strong primary-care tradition and a mixed public and private delivery model, was moving from a national digital-health strategy to an implementable blueprint. Health service delivery is shared between the central Ministry of Health and the nine provincial councils, which meant the design problem was, again, one of federation, not centralisation, with a political economy specific to Sri Lanka’s constitutional arrangement.

The role

Ron served as lead architect and adviser to the ministry programme office, working through the eSHIFT Partner Network. The scope covered:

• the target reference architecture (registries, repositories, indexes, viewer, terminology services);
• the standards profile for cross-jurisdictional interoperability, anchored on HL7 FHIR;
• a staged roadmap aligned to the ministry’s procurement, workforce, and budget cycles;
• capacity-building with the local architecture team so that ownership transferred at engagement end.

The craft

Three decisions worth naming:

1. FHIR first, but profiled hard. FHIR was chosen for its practitioner acceptance and vendor availability, but the useful artefact was the national implementation guide: a tight profile that constrained FHIR to the resources and elements the country actually needed, with terminology bindings to local coding systems.
2. Federated identity as the foundational service. Nothing else in the architecture was allowed to proceed until identity (patient, provider, facility) had a clean, cross-jurisdictional resolution pattern. This ordered the roadmap.
3. Roadmap as procurement plan. Each phase in the roadmap was scoped so that it corresponded to a single realistic ministry procurement, with a defined vendor market and known unit costs. Phases were not defined by technology waves; they were defined by what the ministry could actually contract.

The outcome

The blueprint was accepted by the ministry programme office and became the reference against which subsequent implementation contracts were structured. The local architecture team, trained through the engagement, took over stewardship of the artefact at handover and has extended it since.

What it teaches

The engagement is a case study in adapting a federated pattern to a specific political economy. The core architectural idea (registries and repositories, not warehouses) was portable; the governance, sequencing, and standards decisions had to be rebuilt from scratch for the Sri Lankan context. The harder lesson, visible in the years since, sits at the interface between a sound blueprint and the institutional capacity to procure and engineer against it. That interface is a recurring subject of Ron’s practice.