Description
Large databases of aggregated electronic health record (EHR) data have the potential to efficiently identify rare events. Here we used de-identified EHR data from 2,740,000 breast cancer cases through Observational Health Data Sciences and Informatics collaborative (OHDSI) EHR data sets. The rate of left ventricular dysfunction in breast cancer patients treated with antimetabolites was 2-3 times higher than those with other therapies. OHDSI can efficiently investigate correlations otherwise would be significantly more difficult to assess.
Learning Objective: To provide a case example of how to formulate an approach to apply OHDSI (Observational Health Data Sciences and Informatics collaborative)/OMOP (Observational Medical Outcomes Partnership) CDM (common data model) for efficient large-scale post-market pharmacosurveillance.
Authors:
Lu Pu (Presenter)
School of Medicine, Case Western Reserve University
David Kaelber, Depts of Internal Medicine, Population and Quantitative Health Sciences, The MetroHealth System, Case Western Reserve University
Presentation Materials: