Custom CSS
double-click to edit, do not edit in source
11/15/2025 |
8:30 AM – 12:00 PM |
Room 3
W03: Real-World Analytics with Harmonized Multi-Site EHR Data: An Interactive Tutorial
Presentation Type: Tutorial
Real-World Analytics with Harmonized Multi-Site EHR Data: An Interactive Tutorial
Presentation Time: 08:30 AM - 11:30 AM
Abstract Keywords: Real-World Evidence Generation, Controlled Terminologies, Ontologies, and Vocabularies, Data Mining, Data transformation/ETL
Primary Track: Applications
Programmatic Theme: Clinical Informatics
This 3-hour instructional workshop will introduce attendees, including those with no programming experience, to analysis of multi-site, harmonized medical record and Electronic Health Record (EHR) data. Participants will access a cloud-hosted environment to practice manipulating, analyzing, and visualizing EHR data. We’ll explore graphical and code-based analytic tools and learn about OMOP, a common data model widely used for real-world data research.1 We will discuss the challenges and opportunities real-world data presents in comparison to data specifically produced for research, particularly in the context of multi-site datasets. This includes both novel challenges (e.g. heterogenous missingness) and opportunities (e.g. large-scale machine learning).
This workshop is an updated version of a sold-out 2024 & 2025 AMIA Informatics Summit workshop. The presenters will again use the National COVID Cohort Collaborative (N3C)2 and now also the All of Us (AoU) Researcher Workbench as sample data and analytic platforms to enable participants’ exploration of aggregated patient data as well as row-level synthetic3 OMOP data. N3C hosts a large database of observational EHR records on approximately 23 million patients from 84 data partners4, while AoU hosts data from half a million consented participants from 64 sites. Individual access to the N3C or AoU requires users to be covered by their institutions’ respective Data Use Agreement, which many academic organizations have already signed. The N3C and AoU websites list organizations who have Data Use Agreements in place5. We will focus on mock data developed for training purposes. We will provide facilitators at each table to lead a group exploration, so registration and access to N3C or AoU is optional (but encouraged) for all participants. Instructions for enclave registration are here: N3C - Account Instructions (cd2h.org), and here: All of Us (researchallofus.org) For questions please contact the workshop organizers.
Speakers:
Richard
Moffitt,
Ph.D.
Emory University
Johanna
Loomba,
ME
University of Virginia
Shawn
ONeil,
Ph.D.
University of North Carolina at Chapel Hill
Melissa
Haendel,
PhD
University of North Carolina at Chapel Hill
Eric
Hurwitz,
PhD
University of North Carolina Chapel Hill
Authors:
Andrea Zhou, M.E. - integrated Translational Health Research Institute of Virginia;
Anita Walden, MS - University of North Carolina Chapel Hill;
Margaret Hall, MS - Emory University;
Richard
Moffitt,
Ph.D. - Emory University
Johanna
Loomba,
ME - University of Virginia
Shawn
ONeil,
Ph.D. - University of North Carolina at Chapel Hill
Melissa
Haendel,
PhD - University of North Carolina at Chapel Hill
Eric
Hurwitz,
PhD - University of North Carolina Chapel Hill