BLOG – LAISDAR: Improving COVID-19 Care in Rwanda

edenceHealth is an EHDEN small-to-medium enterprise (SME) based in Kontich (Antwerp), Belgium. We were among the first five SMEs to be certified by EHDEN in September 2019, and we have since had the opportunity to work with many EHDEN Data Partners.

Assisting EHDEN Data Partners in mapping their data to the OMOP Common Data Model has been a key part of our activities these past four years. This work has pushed us to grow as a team and as a company; it has also given us valuable experience and insight that we have brought with us to projects outside the EHDEN network.

One of these outside projects was LAISDAR (“Leveraging Artificial Intelligence and Data Science Techniques in Harmonizing, Accessing and Analysing SARS-COV-2/COVID-19 Data in Rwanda”), which was established in response to the COVID-19 pandemic in Rwanda. The project consortium, led by the University of Rwanda, Ghent University, and the Rwanda Biomedical Centre, was funded by Canada’s IDRC (International Development Research Centre) along with SIDA (Swedish International Development Cooperation Agency) under the Global South AI4COVID Program.

The primary objective of LAISDAR was to create a federated data network of healthcare sites across Rwanda, with sizes ranging from university hospitals to community health clinics. Electronic health record (EHR) data at each site, which includes events like procedures, diagnoses, and prescriptions, was to be augmented with COVID-related data collected at the national level during testing, surveillance, and vaccination efforts by the Rwandan Ministry of Health (MOH). With this network, LAISDAR aimed to support the Rwandan government in monitoring and predicting the COVID-19 burden. The project also set out to study the impact of various public health measures on the pandemic’s evolution, socioeconomic situations of Rwandans, and mental health.

edenceHealth’s role in the LAISDAR project was to design and implement the technical solution, support the data harmonisation of the different source systems, and support the data sites through deployments and trainings. Fourteen healthcare sites were selected for participation in this project, all of which used either OpenMRS or OpenClinicGA as their primary EHR system. In addition, the Extract-Transform-Load (ETL) processes retrieved site-specific data from (1) national COVID-19 testing registry, and (2) a national survey conducted as part of the project, in order to enrich the sites’ core data sets. Each data site’s data were mapped to OMOP, and an API established at a central server hosted by the MOH facilitated encrypted data transfer between the central COVID data repository and the core EHR data.

We installed and configured all software on fourteen Mac Mini computers, and hand delivered a Mac Mini to each healthcare site in the network. We used Docker to package and deploy all the tools on each machine – this included the OMOP database instance (PostgreSQL), Atlas, R/RStudio (w/HADES), Achilles/DQD, and ARES Indexer. In addition, the ETL programme, developed using Python and SQL, was deployed using Docker. In order to manage system updates, apply application updates, and orchestrate processes, we used SimpleMDM, an Apple-focused mobile device management solution.

ARES, and open-source tool designed to monitor network data quality, played a key role in supporting the quality control and validation processes across the LAISDAR network, because it provided a centralised view on real-time quality issues and mapping statuses across the data sites. We also designed and developed a proof-of-concept (POC) solution for creating monthly quality reports with OMOP CDM data, based on the EHDEN CdmInspection R package. We plan to develop this reporting tool further in conjunction with the participating sites, as it has considerable potential to streamline many processes that are currently done entirely by hand.

LAISDAR officially launched at the end of 2020 and concluded in June 2023. In that time, it established infrastructure and mappings that serve as a solid foundation for further research and collaborations based on observational health data in the OMOP CDM format. LAISDAR has also inspired analogous projects, both within Africa and in low-to-middle-income countries (LMICs) elsewhere, that will make use of the solutions and strategies we developed and deployed.

LAISDAR has been both a fulfilling and challenging project. It has expanded our perspective on data harmonisation in many ways, and although it was not linked directly to EHDEN, we feel that our role as an EHDEN SME certainly helped prepare us for a project of LAISDAR’s scope and size.

 

Lars Halvorsen

Managing Partner

edenceHealth

 

For additional information, please also see the EHDEN Community Call #11 recording, and our related poster presentations from the 2023 European OHDSI Symposium:

 

Acknowledgments

  • OHDSI
  • EHDEN
  • Ghent University team, lead by project promoter Dr. Marc Twagirumukiza
  • University of Rwanda team, lead by Principal Investigator Dr. Charles Ruranga
  • Rwanda Biomedical Centre (RBC) team
  • Participating hospitals and their IT staff and data managers
  • edenceHealth team
  • Canada’s International Development Research Centre (IDRC) and the Swedish International Development Cooperation Agency (Sida)
  • Global South AI4COVID Program