Pathogen Dynamics Group - Public Health Data Architecture
Conceptual data flow design for a single integrated platform for clinical microbiology, real-time epidemiology and intervention research to fight infectious pathogens in low income settings
Researchers at the Pathogen Dynamics Group, part of the University of Oxford, are working to develop a scalable laboratory infrastructure, initially in Zambia, for rapid diagnostics and whole genome viral sequencing. The lab will link to digital technologies to broaden access to molecular diagnostics while informing public health efforts aimed at prevention and improved clinical outcomes. Central to the innovation is a bioinformatics pipeline and sequencing method, optimised to acquire key clinical and epidemiological data such as transmission statistics, drug resistance, recency of infection, co-infection, and viral load.
Stakeholders
1 to 1 consultancy support to the lead researcher
Size
c.15 professional days over two months
Project Type
Data design project, conceptual systems
Year
2022
The platform is intended to deliver relevant inferences to clinicians, policy makers and people living with disease, in order to improve clinical outcomes and curtail local outbreaks in areas of Sub-Saharan Africa that are inadequately served by existing infrastructure for clinical microbiology. The system handles extremely sensitive patient data and the team required support in exploring conceptual level data architecture and data flows between patients, laboratories, care settings and public health organisations.
Our team has data flow design experience from both corporate IT environments and from working with small-scale digital products. Critically, this meant we were able to consider and comprehend the intricacies of this particular data challenge. Patient privacy is of paramount importance, but provision of access to sequencing data from multiple individuals introduces new possibilities for making inferences in relation to disease transmission, which might compromise this privacy. To be successful, the system required strong safeguards to prevent retrospective analysis of epidemiological data to make unwarranted clinical observations at an individual level.
We explored a range of conceptual options with the client and developed visual data flow diagrams showing how information should pass (or not) between different data producers and consumers.
Project Retrospective
-
The solution will establish scalable and portable, high-throughput sequencing laboratories in Sub-Saharan Africa, and contribute positively to epidemic early warning systems and management. The project is funded by the Bill and Melinda Gates foundation and is a collaboration with Zambart (LSHTM and University of Zambia's School of Medicine).
-
The project was challenging because of the highly specialist subject matter - although we have a very strong understanding of data management principles and practice, some of the intricacies of epidemiological data and viral sequencing went beyond our comfort zone. We overcame this challenge by working with the client to reduce this complexity through describing and categorising different streams of epidemiological and clinical data, sorting them into broad groupings according to their main downstream uses. This provided a common language to work with that was a simplification of reality, but with enough detail to explore solutions.
-
We delivered value in this project by converting an unstructured, unbounded debate about patient data risks into a conceptual, visual model which could be used as a ‘straw man’ to pursue a more concrete understanding of the problem and potential solutions. While our contribution to this very large, multi-year project was small, our work helped the team to move past a blocker.
-
This short project was a helpful return to technical systems thinking which had been a mainstay of our consultancy work in the past. There were parallels at a high level with our later work for JNCC in the development of new data models for marine pressures and features, and our experience of working with healthcare data systems was relevant when it came to working within NHS England’s IT landscape.