Skip to main content

Data-Driven Approaches to Improve Phenotype Sensitivity Using EHR Data

    Basic Details

    Medical product safety studies traditionally use condition-specific diagnosis codes as filters to identify patients with health outcomes of interest, but such filters may lack sensitivity. With COVID-19 as a use case, this poster identifies surrogate features for these diagnosis codes using electronic health record data (coded procedures, labs, medications, problem lists, and diagnoses) and evaluates whether such surrogates improve sensitivity by identifying cases overlooked by a traditional filter. It was presented at the 38th International Conference on Pharmacoepidemiology and Therapeutic Risk Management. 


    Joshua C. Smith, Daniel Park, Jill Whitaker, Michael F. McLemore, Elizabeth E. Hanchrow, Dax Westerman, Joshua Osmanski, Robert Winter, Saranrat Wittayanukorn, Danijela Stojanovic, Arvind Ramaprasan, Ann Kelley, Mary Shea, David J. Cronkite, Jing Zhou, Yueqin Zhao, Darren Toh, Kevin B. Johnson, David Aronoff, David S. Carrell