About DHS Research Data Sources

Through the Clinical Translational Science Awards Program (CTSA), USC provides research informatics services for USC, UCLA, and other academic partners. The primary source of structured data for research is the DHS ELM Data Warehouse, which harmonizes all clinical data from the Cerner Electronic Medical Record, third-party insurance claims for empaneled patients at DHS sites, and geographically linked data for census-tract level linkages.

Clinical and administrative data from approximately 4.8M patients and 36.8M encounters are curated for analyses, including 126 curated questionnaire types and 180,348 mental health assessments since 2020. Data are updated continuously; currently a 12-24 hour delay between the EMR and research data marts. In addition to serving the Cerner i2b2 instance, the Informatics Core maintains dozens of identified and de-identified data marts for research purposes. Data conform to FHIR® terminology and modeling standards, and can be readily transformed into common data models for research, including the OMOP Common Data Model and the PCORnet Common Data Model.