Abstract
Purpose :
Standardization of ophthalmic data and imaging is an ongoing process led by the American Academy of Ophthalmology Data Standards Workgroup. In 2022, standardization efforts expanded to include the OMOP common data model (CDM), which is used in NIH data generation projects such as All of Us and Bridge2AI. Although the OMOP CDM has been in existence for over a decade, it does not yet include ophthalmic concepts. The Eye Care and Vision Research Workgroup in The Observational Health Data Science and Informatics (OHDSI) community was created to address these gaps.
Methods :
The OHDSI Eye Care and Vision Research Workgroup was formed in June 2022 to perform a comprehensive mapping analysis of data elements, identify gaps in standardized terminologies, and identify motivating use cases that would prioritize which concepts to add to the OMOP CDM. Subgroups focusing on specific sub-specialties of ophthalmology were formed to allow for concurrent efforts to accomplish mapping tasks. The workgroup disseminated work in publications, in person meetings at vision research conferences, presentations, and short courses to foster interest in the project.
Results :
In the first year, the workgroup met 17 times and grew to 122 members, most of which were clinician-scientists and informaticians. Six subgroups formed to focus on ophthalmic sub-specialties and different work tasks. Together, the group attempted to map 3700 ophthalmic data concepts to standardized terminologies. For example, we found roughly 75% of eye exam elements in Epic electronic health record did not have a precise matching concept in OMOP or other standardized terminologies (Fig 1). Missing terms were submitted to SNOMED CT and LOINC. Eighteen phenotypes for ophthalmic disease were developed. The group published 15 manuscripts and presented 18 talks and 5 posters. Fig 2 shows a timeline and highlights milestones of the year.
Conclusions :
The first year of the OHDSI Eye Care and Vision Research workgroup was highly productive and laid the groundwork for future efforts. The goals for the second year are to continue mapping concepts, integrate imaging, and add ophthalmic data to OMOP datamarts at multiple institutions.
This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.