Design and implementation of FAIR principles in digital health data in Uganda, Africa

Doctoral Thesis

This thesis investigates the design and implementation of the FAIR principles—Findable, Accessible, Interoperable, and Reusable—to strengthen digital health data management in Africa, with Uganda as the primary case study. In Uganda, substantial volumes of health data remain paper-based and are subsequently aggregated into DHIS2, limiting their accessibility, interoperability, and reuse by healthcare professionals, researchers, and other stakeholders.

Situated within the VODAN-Africa and Asia initiative, the research examines health data practices, regulatory frameworks, digital health infrastructures, and FAIR-enabled architectures that support secure, machine-actionable, and locally governed data. The study evaluates digital health platforms, applies FAIR Data Points, and demonstrates the VODAN-in-a-Box architecture for privacy-preserving access without requiring data to leave healthcare facilities.

The thesis further compares FAIR-based approaches with DHIS2 and...Show moreThis thesis investigates the design and implementation of the FAIR principles—Findable, Accessible, Interoperable, and Reusable—to strengthen digital health data management in Africa, with Uganda as the primary case study. In Uganda, substantial volumes of health data remain paper-based and are subsequently aggregated into DHIS2, limiting their accessibility, interoperability, and reuse by healthcare professionals, researchers, and other stakeholders.

Situated within the VODAN-Africa and Asia initiative, the research examines health data practices, regulatory frameworks, digital health infrastructures, and FAIR-enabled architectures that support secure, machine-actionable, and locally governed data. The study evaluates digital health platforms, applies FAIR Data Points, and demonstrates the VODAN-in-a-Box architecture for privacy-preserving access without requiring data to leave healthcare facilities.

The thesis further compares FAIR-based approaches with DHIS2 and explores the integration of machine learning with FAIR health data. Using anonymized outpatient and antenatal datasets from Uganda, machine learning models are applied to prediction, anomaly detection, and clustering tasks.

Overall, the findings demonstrate that FAIR principles provide a viable framework for enhancing interoperability, data reuse, privacy-preserving analytics, and evidence-based decision-making. The research offers practical and policy-oriented recommendations for strengthening digital health ecosystems in Uganda and comparable settings, while supporting sustainable, collaborative, and equitable use of health data across Africa.
Show less

Data management, Information flow

FAIR principles, Hospital information system(HIS)

Electronic medical record (EMR)

Electronic health record (EHR)

Patient health record (PHR)FAIR data, machine learning

Federated learning

Federated architecture

Outpatient register

Antenatal care register

Allegrograph triple store