All case studies
Integration workHealth systems · Data integration

Building health-data pipelines for national EHR interoperability

Structured data exchange and reporting workflows connecting neonatal care data with national health information systems.

My role
Software Developer & Technical Consultant
Period
2025–Present

PROJECT OVERVIEW

This work supports structured exchange between Neotree, the Impilo electronic health record and DHIS2 reporting workflows in collaboration with health-sector stakeholders in Zimbabwe.

The problem

Independent health platforms use different data models, identifiers and operational rhythms. Exchanging data safely requires more than connecting APIs: records must be mapped, validated, traceable and recoverable when a downstream system is unavailable.

My responsibility

I contribute to integration design and implementation, validation, error handling and reporting pipelines. The architecture shown here is intentionally high-level and uses no patient, facility or internal infrastructure data.

REAL-WORLD CONSTRAINTS

Sensitive clinical data and strict privacy boundaries
Different schemas and identifiers across systems
Partial failures and delayed delivery
Auditability and data-quality requirements
Multiple organisational stakeholders

SOLUTION & CONTRIBUTION

The integration pipeline validates and maps structured records before controlled delivery to downstream systems. Explicit processing states, retry-safe operations and observable failures reduce ambiguity and make reconciliation possible without exposing clinical details.

  1. 01Implemented structured mapping between platform and reporting data models.
  2. 02Built validation paths that reject or flag incomplete records before delivery.
  3. 03Contributed to retry and error-handling behaviour for downstream outages.
  4. 04Supported secure API integration and reporting workflows.
  5. 05Collaborated with technical and health-sector stakeholders on interoperability requirements.

SANITISED ARCHITECTURE

Neotree records
Validation and mapping
Secure integration services
Impilo / DHIS2

Certain implementation details and screens are intentionally anonymised to protect organisational and user data.

Engineering challenge

Integration failures are rarely all-or-nothing. A defensible pipeline needs to distinguish invalid source data, transient transport errors and downstream rejections, then preserve enough context for safe replay and audit without duplicating accepted records.

Observable outcomes

  • Structured pathways for health-data exchange
  • Clearer validation and failure states
  • Support for national reporting and interoperability workflows
  • A reusable approach to integration reliability

Technology used

REST APIsNode.jsTypeScriptPostgreSQLDHIS2EHR integrationDocker

Next case study

Application update and device management