A scoping review of exercise oncology trials to inform best practice recommendations for exercise in older adults (65+ years) living with and beyond cancer.

The extensive use of electronic health records (EHR), picture archiving and communication systems (PACS), and health information technology infrastructure has produced a vast amount of digital data for analysis. However, secondary use and harmonization of these data are hampered by the incomplete representation of data elements in standardized medical terminologies in certain areas, particularly for specialized domains like ophthalmology. As a result, there is limited availability of real-world data for research and interoperability. Specifically, in the field of ophthalmology and glaucoma, we have shown a significant lack of concept representations in standardized terminologies. These standardized terminologies are often incorporated into common data models (CDMs), which are frameworks designed to harmonize and standardize data silos and “big data” resources by imposing a uniform structure and vocabulary, enabling seamless integration and analysis of disparate data sets. Clinical ophthalmology practice routinely monitors vision-related quality of life by standard automated perimetry (SAP). Given its frequent use in eye care, a DICOM standard for SAP tests was created, which includes all the data elements included by SAP vendors on single test reports. The DICOM supplement for Ophthalmic Visual field (OPV) has been available since 2006 ( for further details). Despite the widespread use of SAP, the significance of the test in eye care, and the availability of a DICOM standard, SAP data are often unavailable in “big data” resources (eg, NIH’s , institutional and multi-institutional EHR data warehouses, and specialty data registries). This may be attributed to the existence of these data outside the EHR, the difficulty of extracting these data from the devices and vendor-specific image management systems, and limited representation in the Observational Medical Outcomes Partnership (OMOP) CDM, which increasingly forms the basis for clinical research data resources. However, the extent of gaps in representation has not been quantified previously. Further, this hampers other efforts based on standardized data elements, including but not limited to Health Level Seven (HL7) Fast Healthcare Interoperability Resources (FHIR) implementation guides; this limits the ability to exchange data between different repositories. The OMOP CDM has been designed to harmonize the representation of observational health data. Recently, it has been extended to include 2 new tables, “image_occurrence” and “image_features,” to support imaging research and outcome studies utilizing imaging features. This extension facilitates the inclusion of medical image metadata in OMOP and accommodates links to the images themselves. Given the existing functionality of OMOP, this will enable new research questions related to medical imaging (and testing) and collaborations across institutions. Identifying gaps in data element representation will help inform what concepts are needed to enable the representation of SAP data in the OMOP CDM and, therefore, facilitate their future inclusion in data warehouses for multi-institutional research initiatives. Recognizing this need for data standards around ophthalmic imaging data, we analyzed gaps in SAP data coverage in the OMOP CDM and standardized terminologies in the current study. We also propose new codes to be added to Logical Observation Identifiers Names and Codes (LOINC), establishing a foundation for future data transformations from various SAP devices into the OMOP CDM.