2023 OHDSI Symposium

Oct. 20-22 • East Brunswick, New Jersey

The 2023 OHDSI Global Symposium welcomed more than 440 of our global collaborators together for three days of sharing research, forging new connections and pushing forward together the OHDSI mission of improving health by empowering a community to collaboratively generate the evidence that promotes better health decisions and better care. 

This page will be home to all materials from the global symposium. #JoinTheJourney #OHDSI2023

State of the Community

Various leaders within OHDSI shared a presentation on the state of the community, with specific focuses on data standards, vocabulary enhancements and open-source development. Speakers included:

George Hripcsak, Columbia University
Clair Blacketer, Johnson & Johnson
Alexander Davydov, Odysseus Data Services
Katy Sadowski, Boehringer Ingelheim
Peter Rijnbeek, Erasmus MC
Mengling ‘Mornin’ Feng, National University of Singapore

Plenary: Improving the reliability and scale of case validation

Case validation is regarded as a necessary element of regulatory-grade evidence, but conducting case validation through human adjudication of source records is time- and resource-intensive, has unknown performance, and is frequently conducted in such a way that does not enable either full caseset review or proper quantitative bias analysis.  In this plenary, OHDSI collaborators presented innovative methodological research and open-source development to improve the reliability and scalability of the case validation process, demonstrating that it may be possible to replace source records through an informatics-enhanced patient profile of structured data from the OMOP CDM (KEEPER), and to supplement human review through the use of large language models to estimate measurement error and identify 

differential misclassification. KEEPER + LLM was empirically evaluated in 10 diseases across 3 experiments in 2 different data sources, and revealed that there can be substantial heterogeneity in agreement between human reviewers but that LLMs agree with humans as much as humans agree with each other. Speakers included: Patrick Ryan, Johnson & Johnson, Columbia University; Anna Ostropolets, Odysseus Data Service; and Martijn Schuemie, Johnson & Johnson, University of California, Los Angeles

Panel: Lessons learned from OHDSI network studies

Network studies are challenging, but they are provide a critical opportunity to generate robust, reliable real-world evidence. Leaders from recent OHDSI network studies share insights, challenges and possible solutions from recent network studies, including from the SOS Challenge. Speakers included:

Marc Suchard, University of California-Los Angeles
Cindy Cai, Johns Hopkins University
Seng Chan You, Yonsei University
Anthony Sena, Johnson & Johnson

Moderator: Sarah Seager, IQVIA

Collaborator Showcase Lightning Talks

There were more than 170 submissions for the 2023 Collaborator Showcase, and 10 posters were selected to present lightning talks during Friday’s portion of the symposium.

Session I

1) Mapping of Critical Care EHR Flowsheet data to the OMOP CDM via SSSOM • Polina Talapova, SciForce
2) Paving the way to estimate daily dose in OMOP CDM for Drug Utilisation Studies in DARWIN EU® • Theresa Burkard, University of Oxford
3) Generating Synthetic Electronic Health Records in OMOP using GPT • Chao Pang, Columbia University
4) Comparing concepts extracted from clinical Dutch text to conditions in the structured data • Tom Seinen, Erasmus MC
5) Finding a constrained number of predictor phenotypes for multiple outcome prediction • Jenna Reps, Johnson & Johnson

Moderator: Davera Gabriel, Johns Hopkins University

Session II

1) Synthesizing Evidence for Rare Events: a Novel Zero-Inflated Bivariate Model to Integrate Studies with Double-Zero Outcomes • Lu Li, University of Pennsylvania
2) Active Safety Surveillance Using Real-world Evidence (ASSURE): An application of the Strategus package • Kevin Haynes, Johnson & Johnson
3) Patient’s outcomes after endoscopic retrograde cholangiopan creatography (ERCP) using reprocessed duodenoscope accessories: a descriptive study using real-world data • Jessica Maruyama, Precision Data
4) Does COVID-19 Increase Racial/Ethnic Differences in Prevalence of Post-acute Sequelae of SARS-CoV-2 infection (PASC) in Children and Adolescents? An EHR-Based Cohort from the RECOVER Program • Bingyu Zhang, University of Pennsylvania
5) Eye Care and Vision Research Workgroup: First Year Update • Michelle Hribar, National Institutes of Health – National Eye Institute

Moderator: Atif Adam, IQVIA

Collaborator Showcase Posters & Software Demos

We received a record number of submissions for the 2023 Collaborator Showcase, following detailed review by community volunteers in the Scientific Review Committee, there were 137 posters and 24 software demos that were presented during the collaborator showcase.

Please visit the link below to visit the posters, brief reports and other supplementary materials for each showcase submission. Each submission will be profiled in the #OHDSISocialShowcase, so please make sure you follow OHDSI on Twitter/X, LinkedIn and Instagram

Tutorial: Introduction to OHDSI

The journey from data to evidence can be challenging alone but is greatly enabled through community collaboration. In this half-day tutorial, we will introduce newcomers to OHDSI. Specifically, about the tools, practices, and open-science approach to evidence generation that the OHDSI community has developed and evolved over the past decade. 

Faculty will highlight the ways community individuals can participate as well as receive value from the community’s outputs. The course will include topics such as open community data standards – including the OMOP Common Data Model and OHDSI Standardized Vocabularies, opensource analytic tools – including HADES and ATLAS, and the conduct of open network studies for methodological research and clinical applications.

AGENDA
8:00-8:15 • Introduction to OHDSI
8:15-9:00 • Data Standardization
9:00-9:45 • Methodologic Research

10:00-10:15 • Break
10:15-11:00 • Open-Source Development
11:00-11:45 • Clinical Evidence Generation
11:45-12:00 •  Conclusion and Next Steps

Closing & Titan Awards

The closing of Day 1 of the OHDSI Global Symposium turned into an experiment in large-scale collaboration … in the form of a massive escape room inside the Hilton East Brunswick Hotel & Executive Meeting Center! Check back later for a video recap of this fun and interactive activity!

Prior to the closing talk, the 2023 Titan Award winners were announced. Congratulations to our honorees, who were both nominated and selected by peers in the community:

Data Standards: Gowtham Rao and Azza Shoaibi
Methodological Research: Jiayi (Jessie) Tong
Open-Source Development: Katy Sadowski
Clinical Applications: Center for Surgical Science
Community Leadership: Nicole Pratt
Community Collaboration: Cynthia Sung
Community Support: Gyeol Song

Top