Development of a COVID-19 Application Ontology for the ACT Network

Area of Science:

• Biomedical Informatics
• Health Services Research

Background:

• Electronic Health Records (EHRs) are vital for COVID-19 research, but data harmonization across institutions is challenging.
• Multisite research requires standardized data to effectively analyze large patient populations.

Approach:

• Developed a COVID-19 application ontology within the Accrual to Clinical Trials (ACT) network.
• The ontology includes over 50,000 concepts covering diagnoses, procedures, medications, and lab tests.
• Incorporated computational phenotypes for illness course and outcomes, and harmonized SARS-CoV-2 lab test values.

Key Points:

• The ontology facilitates the harmonization of critical data elements for COVID-19 research.
• It was successfully deployed and validated on the ACT COVID-19 network, encompassing data from 14.5 million patients across nine academic health centers.
• Includes specific computational phenotypes and harmonized value sets for SARS-CoV-2 laboratory tests.

Conclusions:

• The developed COVID-19 ontology enables seamless data harmonization for multisite EHR research.
• This freely available resource (on GitHub) will enhance COVID-19 research beyond the ACT network.
• Facilitates standardized analysis of EHR data for improved understanding of coronavirus disease 2019.