The integrated Translational Health Research Institute of Virginia (iTHRIV): Using Data to Improve Health

SARS-CoV-2大流行的未知和不断变化的特征严重挑战了国际卫生组织

The unknown and changing characteristics of the SARS-CoV-2 pandemic have severely challenged the United States (U.S.) health care systems. The key to addressing many of these challenges is data and information sharing. To do this requires bringing together individual level health data from disparate systems into a common structure that can be analyzed for answers to the important questions about COVID-19. Within the health informatics community there are two approaches to integrating data for analysis: (1) Federated data sharing which keeps the data at individual locations and allows for aggregated queries and (2) Harmonized repository that joins the data from the different sites into one database with a common data model that allows for individual or row level queries. While the federated approach is easier to implement and much more widely used, the harmonization approach is what is needed to address the challenges of the COVID-19 pandemic since it will enable more impactful data analysis on the scientific questions surrounding this disease. The University of Virginia (UVA), the lead site for the cross-state integrated Translational Health Research Institute of Virginia (iTHRIV), is well positioned to serve as an initial, pilot provider of data for the harmonized, analytic database being assembled by the National Center for Advancing Translational Sciences (NCATS) known as the National COVID Cohort Collaborative (N3C). There four reasons iTHRIV can do this at UVA: 1) iTHRIV has implemented the Observational Medical Outcomes Partnership (OMOP) Common Data Model (CDM) and this is not only the accepted CDM for data transfer to N3C but it also the target data transfer model for N3C, which will make the iTHRIV CDM a good choice to validate data transforms; 2) The iTHRIV informatics team have been active participants in the development of the COVID-19 Phenotype implementation in OMOP and we can thus quickly implement the data queries; 3) The iTHRIV data Commons utilizes an architecture which includes multiple CDM and this gives us the capability to expand data acquisition to all partner institutions in iTHRIV and to rapidly respond to changes required in data acquisition and transfer; and 4) The University of Virginia has an IRB Reliance Agreement in place with SMART IRB and can rely on any non-UVA IRB that also has an IRB Reliance Agreement with SMART IRB, which will streamline our start-up process for participation. iTHRIV at UVA therefore provides an ideal pilot site for the N3C project, and brings the iTHRIV Commons and the iTHRIV partners institutions to rapidly support rapid expansion to other CDM as a model for the larger consortium. The Commons also provides a leading team-science platform during the follow-on phases of N3C where researchers within Virginia can collaborate with others from around the U.S. and the world to analyze the data collected in centralized repository by the N3C project and address impactful health problems for the community.