Programmatic Theme: Clinical Research Informatics

Abstract: The universal adoption of electronic health records presents an unprecedented opportunity to fuel population-scale development of research-grade computational phenotypes (CPs). However, several barriers to the development, validation, and implementation of CPs must be overcome before their potential can be fully realized. We provide a comprehensive examination of how different vocabulary mapping strategies, across multiple clinical domains, effects the creation of patient cohorts for seven different CPs.

Learning Objective: After participating in this session, the learner should be better able to:

1. Understand how current barriers to deriving computational phenotypes that prevents the development of automated computational methods.

2. Learn about the trade-offs in terms of error and information loss vs. reduced clinical code sets and information gain when employing different vocabulary mapping strategies as part of aligning pre-built computational phenotypes to the OMOP common data model.

3. Hypothesize explanations for the large differences in cohort size observed when creating patient cohorts using only the clinical code sets versus the clinical code sets (as well as when varying the included clinical domains) in addition to the phenotype definitions or clinical logic.


Tiffany Callahan (Presenter)
University of Colorado Denver

Jordan Wyrwa, University of Colorado Denver Anschutz Medical Campus
Katy Trinkley, University of Colorado Skaggs School of Pharmacy and Pharmaceutical Sciences
Lawrence Hunter, University of Colorado Denver
Michael Kahn, University of ColoradoDenver Anschutz Medical Campus
Tellen Bennett, University of Colorado Denver

Keywords, Themes & Types